Je viens du monde de l’ingénierie et de la recherche, où la structure, la précision et la clarté priment avant tout. Cependant, j’ai toujours été attiré par les grands conteurs (romanciers, vulgarisateurs d’histoire, humoristes, communicateurs scientifiques). Il y a quelque chose de presque magique dans leur capacité à tisser des idées pour en faire des récits captivants. Ma propre écriture a toujours été plus pragmatique : spécifications détaillées, notes internes synthétisant de multiples perspectives, documents exposant une vision, articles de recherche partageant les résultats d’expériences, etc. Au cours des vingt dernières années, j’ai appris que la communication écrite détient un pouvoir particulier : celui de façonner et de propager les idées, tant dans mon propre esprit que dans celui des autres.

Pour moi, l’écriture n’a jamais été une activité confortable, car j’écris pour réfléchir. Écrire oblige mon esprit à se confronter à la réalité : saisir de multiples points de vue, se forger une opinion, articuler une perspective. Sans l’écriture, mes idées restent vagues, mal formulées et peu étayées.

Je partage souvent de courts textes avec un collègue pour qu’il y jette un coup d’œil rapide, et il m’arrive parfois de les relier entre eux pour publier un essai, structurer un cours ou faire une présentation orale. Cependant, la plupart du temps, j’accumule des piles de brouillons inachevés qui me servent de notes pour nourrir ma réflexion. Jusqu’à récemment, il s’agissait surtout d’un travail solitaire. C’est toujours le cas, fondamentalement. La réflexion, le choix et l’élaboration des idées restent mon affaire. Mais l’acte d’écrire, l’articulation elle-même, a changé.

L’écriture devient conversationnelle

Avec l’avènement des outils d’IA générative, certaines étapes de mon processus d’écriture se déroulent désormais via une interface de messagerie. Certains parlent de « vibe writing ». Du jour au lendemain, j’ai pu engager un dialogue autour de mes idées, tester différentes façons de les formuler et obtenir de l’aide lorsque je me sentais bloqué. Pour quelqu’un qui a du mal à raconter des histoires, cela m’a semblé révolutionnaire. Je disposais d’un outil pour m’aider à transformer le flux chaotique de mes pensées en textes fluides et lisibles, presque comme un conteur.

Je m’appuie sur des LLM polyvalents populaires (par exemple ChatGPT et Claude) pour affiner le fil de mes pensées, et sur des wrappers IA (par exemple Perplexity) pour trouver des références et de nouveaux documents de recherche. J’utilise ces outils pour organiser mes notes, définir les limites de mon texte et mieux positionner mes idées. Ils m’aident particulièrement à éviter de « Magnum Opuser » : ce piège courant où la portée du projet s’étend à l’infini et où les notes s’accumulent jusqu’à ce que le projet devienne trop complexe pour être jamais achevé.

Si je devais illustrer mon processus d’écriture pour ce texte, cela ressemblerait à une ligne sinueuse qui prendrait peu à peu forme. La ligne bleue montre comment le texte a évolué, tel un document vivant, au fil de révisions successives, passant de fragments vagues à des idées plus claires et mieux développées. Les lignes gris clair représentent des notes issues d’observations, de lectures et de conversations qui croisent les formulations générées par l’IA, en rose. Ensemble, elles tracent un parcours exploratoire qui se divise en plusieurs directions, revient sur ses pas et finit par converger. Ce qui commence par un champ d’étude restreint s’élargit pour devenir non seulement un texte achevé, mais aussi un paysage de pensée plus vaste qui n’émerge qu’à travers l’acte d’écrire.

Ce schéma reflète mes autres travaux créatifs dans les domaines de l’ingénierie logicielle et de la conception prospective, où les outils d’IA générative ont élargi mes capacités. J’ai pu concrétiser des idées et prototyper des concepts sans avoir une connaissance approfondie des dernières bibliothèques logicielles ni maîtriser certains langages de programmation.

Après avoir passé des années à créer des logiciels et à imaginer l’avenir, j’ai appris à rester attentif à ce que nous risquons de perdre au profit de ce que nous gagnons. En 1964, Marshall McLuhan affirmait dans “Understanding Media” que la technologie et la société coévoluent : toute amélioration est aussi une perte.

Les coûts cachés de la rédaction assistée par l’IA

Les grands modèles de langage (LLM) ont sans aucun doute réduit les obstacles à l’écriture. Beaucoup de gens peuvent désormais exprimer leurs pensées avec plus d’assurance, même lorsqu’ils écrivent dans une langue étrangère. Une chose qui aurait été inimaginable il y a seulement quelques années. Nous nous soucions moins de la grammaire et de la forme, car celles-ci semblent « facilement » corrigibles par un outil d’écriture basé sur l’IA. Ou s’agit-il d’une illusion de confiance ?

Perte d’authenticité

Mon expérience de l’externalisation de la rédaction aboutit souvent à un style prudent et prévisible, au lieu de repousser les limites de ma pensée originale. Si je ne fais pas attention, mes idées s’estompent et deviennent banales. Elles sont dépouillées de leurs nuances personnelles, perdent leur wabi-sabi et finissent rapidement par manquer d’authenticité. Lorsqu’on l’aborde de manière superficielle, l’écriture assistée par l’IA peut facilement réduire la réflexion à un juste milieu statistique.

Raisonnement erroné

Les outils d’IA générative se précipitent vers des conclusions et des solutions « suffisantes », introduisant parfois des raccourcis illogiques. Ils produisent une prose plausible, convaincante et assurée qui peut camoufler un raisonnement erroné. Ces biais sont des caractéristiques des modèles probabilistes stochastiques utilisés dans l’IA générative. Ce ne sont pas des bogues qui disparaîtront de sitôt. Par conséquent, je me retrouve engagé dans un effort constant pour me réapproprier les idées, pour les faire à nouveau miennes.

Pensée superficielle

Enfin, la pensée n’est pas linéaire comme une conversation via une interface de messagerie. Elle est souvent erratique et désordonnée. Pour développer des idées plus solides, il est nécessaire de s’exposer à différentes perspectives et de faire preuve de discipline pour les approfondir. Les idées ont besoin de temps pour mûrir. Les outils d’IA générative actuels sont indéniablement utiles, mais leurs usages reflètent également la culture du raccourci d’aujourd’hui, nous incitant à éviter le travail intellectuel difficile requis pour des pensées véritablement originales et profondes.

Si je devais illustrer mon expérience de l’écriture entièrement assistée par l’IA, le schéma raconterait cette histoire : la trajectoire évolue rapidement vers une clarté apparente et une « illusion de maturité », à mesure que les outils d’IA m’aident à lisser les phrases, à resserrer la structure et à définir les limites. Cependant, l’étendue du corpus d’écriture reste relativement limitée. Le texte final semble complet et abouti, mais il reflète ce que je perds : la réflexion plus profonde et plus erratique qui naît de la confrontation avec des idées en dehors du contexte d’une interaction avec une machine probabiliste.

Au-delà de mon travail, j’ai observé cette « illusion de maturité » dans les projets finaux de certains étudiants et dans les contenus générés par l’IA partagés sur LinkedIn, Medium et les blogs. Des études récentes semblent corroborer ces observations, bien que la recherche en soit encore à ses débuts et que ces résultats doivent être considérés avec prudence. La cause sous-jacente semble être notre préférence humaine naturelle pour la réflexion rapide, qui est plus facile et nécessite moins de ressources.

Cela soulève un défi crucial en matière de conception : comment maintenir en vie le travail de réflexion, lent et ardu, alors que les outils d’IA générative rendent l’écriture si rapide et si facile ?

L’écriture est une activité sociale

Pour rédiger ce texte, j’ai souhaité contrebalancer mon utilisation d’outils d’IA générative par davantage d’interactions humaines. J’ai mené une petite expérience sous la forme d’une tertulia. Concept originaire d’Espagne, la tertulia est un rassemblement social régulier où les participants partagent leurs dernières créations et discutent de l’actualité. Ces rassemblements, également appelés « cénacle » en France ou « salon » dans le monde anglophone, constituent depuis longtemps des espaces où les idées s’épanouissent grâce aux liens humains.

Concrètement, une fois par semaine, je réunissais 4 à 6 collègues (que l’on appelle les « tertulianos ») pour une discussion en ligne d’une heure. Chacun d’entre nous apportait un projet en cours, tel qu’un brouillon, un projet, un plan de présentation, des lectures, etc. Nous utilisons tous régulièrement des outils d’IA, mais la tertulia est devenue un espace de réflexion où nos idées pouvaient mûrir loin de l’agitation du travail.

Ces sessions donnaient l’impression de participer à un « cercle d’écriture », un luxe que seuls les conteurs ou les humoristes populaires ont le privilège de s’offrir. Nous échangions des idées, nous nous remettions mutuellement en question et proposions des perspectives auxquelles aucun d’entre nous n’aurait pu parvenir seul ou à l’aide d’un outil. Mon rôle consistait à encourager ces frictions, à provoquer des collisions et à maintenir la conversation suffisamment stimulante pour susciter de nouvelles réflexions. C’était imprévisible. C’était plaisant.

Après chaque session, je me retrouvais plongé dans les notes de nos conversations, ainsi que dans des observations et des lectures connexes. Chaque session a remis en question et approfondi ma réflexion sur « l’écriture pour penser ». J’ai eu du mal à rédiger ce texte même, et cette effort me semblait nécessaire. J’ai pu développer mes idées au-delà de « l’illusion de la maturité ».

Le cœur du processus d’écriture, les idées inattendues, cette sensation de se perdre, les références culturelles obscures, les analogies inspirantes et les véritables élans d’imagination ont tous trouvé leur source dans les interactions sociales. Laurent a évoqué le défi permanent que représente la réappropriation des idées issues des outils d’IA. Andrés a présenté le concept de « culture du raccourci » développé par Carolina Sanín. Lisa a établi des liens avec l’ouvrage Thinking, Fast and Slow. Mes notes regorgent de ces percées, chacune suscitant de nouvelles connexions dans mon esprit.

Un espace de réflexion

Ce texte est le fruit direct de la pratique qu’il décrit : engager des conversations avec des machines pour gagner en concentration et en clarté, et avec des humains pour atteindre maturité et profondeur. L’objectif n’est pas de remplacer la difficulté inhérente à l’écriture, mais de la rendre plus féconde. Il soulève une question plus large sur la manière dont nous gérons collectivement notre relation avec les outils d’IA.

Il existe une citation qui résume bien cette coévolution actuelle entre les machines et les humains :

La diffusion rapide de l’IA est rendue possible par la collaboration humaine.

Elle vient de mon amie et complice Lisa Gansky, membre de la tertulia. Entrepreneure en série et autrice de The Mesh, Lisa sait de quoi elle parle. Elle est experte en technologie, en collaboration et en réseaux. Ensemble, nous partageons une même préoccupation : lorsque la rapidité remplace la profondeur, quelque chose se perd de ce qui fait notre humanité, tant en tant que professionnels qu’en tant que citoyens.

Une réflexion profonde et authentique exige du temps, de la curiosité, de la vulnérabilité et la volonté de s’attarder sur les questions. Comme Lisa le dit souvent, ce n’est pas un « spectator sport » (du sport spectacle). La réflexion implique une création active (par exemple, écrire, dessiner, créer des prototypes, etc.) pour développer et mettre en pratique ses compétences fondamentales.

L’expérience des tertulias montre que tout le monde peut en tirer profit. Mais le monde frénétique de l’IA d’aujourd’hui a besoin de plus que de simples expériences ponctuelles. Il a besoin d’une communauté de pratique pour ce type de réflexion lente et réfléchie. C’est exactement ce que Lisa et moi poursuivons avec Próximo Lab : un espace où une communauté d’apprenants permanents issus de divers horizons s’immerge et interagit les uns avec les autres à travers des tertulias et d’autres explorations collaboratives (par exemple, des ateliers pratiques, des conférences d’invités, etc.). Nous considérons ce type de laboratoire d’apprentissage diversifié et de confiance comme un élément fondamental de nos vies.

Texte traduit par l’auteur. Version originale en anglais:

GIRARDIN, Fabien. Writing with probabilistic machines. Mosaic [online], January 2026, no. 206. ISSN: 1696–3296. DOI: https://doi.org/10.7238/m.n206.2517

Egalement disponible en espagnol:

GIRARDIN, Fabien. Escribir con máquinas probabilísticas. Mosaic [en línea], enero 2026, no. 206. ISSN: 1696–3296. DOI: https://doi.org/10.7238/m.n206.2517

Et catalan

GIRARDIN, Fabien. Escriure amb màquines probabilístiques. Mosaic [en línia], gener 2026, no. 206. ISSN: 1696–3296. DOI: https://doi.org/10.7238/m.n206.2517

Un grand merci à Laurent Bolli, Eva Fernández García, Lisa Gansky, Daniel Goddemeyer, Rohit Gupta, Vytas Jankauskas, Andrés Ortiz et Simone Rebaudengo. Ce texte est le fruit de nos échanges. Je remercie tout particulièrement Andrés Ortiz de m’avoir poussé à me remettre en question et de m’avoir aidé à esquisser un schéma concret de ma nouvelle pratique d’écriture. À la mémoire de Nicolas Nova, avec qui j’aurais tant aimé discuter de tout cela.

By Fabien Girardin and Lisa Gansky. Part of a series on personal software: an introduction, the makers (coming soon), and a planting guide (coming soon) for those who want to make it grow.

These pieces are observations and analysis from the field: a map of the territory and an opening for the conversations that follow. Many of those conversations take place at Pre-Alpha Club.

Software design is in more hands than ever. For most of its history inside organizations, software was something people used rather than shaped — systems procured, platforms rolled out, tools embedded into daily work. An idea rendered by a small group. Everyone else adapted to it or tolerated the gaps. Most people hacked around the parts that didn’t fit. Since the Generative AI jolt of 2024, that arrangement has begun to loosen in genuinely practical ways. The ability to turn a few sentences into functioning software without writing a single line of code is no longer a developer’s privilege. It belongs to anyone who can describe a problem clearly enough for a Large Language Model (LLM) to help solve it. This shift isn’t arriving at the edge of our communities or economy. It’s happening inside regulated banks, global manufacturers, massive consultancies and government institutions. When a multinational bank can enable thousands of employees to shape their own tools without setting their compliance requirements on fire, the risk calculation changes now and for everyone. The question is no longer, is this possible? The pointed question for any leader to ask is: what’s our excuse?

This chapter is about what that shift means for organizations. It explores how personal software — tools built close to the problem, shaped by the people who live inside the work — can leap from being an individual practice to a central organizational capability. It looks at what makes this transition work, what are the obstacles, and what are the essential ingredients embraced by the institutions willing to try.

The core frame here is simple: for decades, organizations faced two options when they needed software. Build it or buy it. Both leave a long tail of specific, local, human-scale needs unserved and almost certainly feel outdated faster than we can say ROI. A third option is now available: Enable.

Built by the People Who Know the Work

In 2024, that balance began to shift (see Software Gets Personal: An Introduction). When given agency and supportive governance frameworks, people who understand their work in detail can translate that understanding into tools that support it, often quickly and without asking for permission. These tools tend to be provisional, and closely tied to a specific context. They’re shaped by proximity to the problem, responsiveness to context, and a sense of completion that comes from temporal fit rather than scalable finish. They get revised as conditions change and shared when they prove useful. Importantly, over time, they mirror how the organization actually functions, rather than how it describes itself in diagrams and airbrushed presentations.

For most large organizations, this kind of activity has long been constrained by organizational design and operating models. Enterprise IT functions evolved to protect stability, security, and coherence at scale. Their core work involved setting boundaries: standardizing tools, controlling access, preventing work from drifting outside prescribed systems. That posture has been a site of struggle between teams and IT in almost every organization we’ve ever had the pleasure of knowing. It made sense to restrict access in a world where software was difficult and expensive to build, fragile to maintain, and risky to improvise. As an unintended consequence, it also trained all of us in organizations to treat deviation as something to suppress or tolerate, rather than something to identify, observe and learn from.

As they grow, organizations develop an immune system. It exists to protect the organism. The mistake many institutions make is confusing protection of the organism with protection of past processes. An immune system that calcifies around yesterday’s operating model eventually protects the wrong thing. If leaders are committed to the future health of the organization, then the task is not to defend familiar structures, but to understand what thriving requires under new conditions. Personal software surfaces this distinction quickly. It reveals whether the immune system still protects the organism — or has begun protecting its own past.

A Culture Question, Not a Technology Question

For organizations bold enough to embrace this shift, the opportunity extends far beyond better tools. At its core, it means building resilience in the face of rapid, global change. The same cultural values that welcome learning and experimentation deliver a more connected, engaged, and responsive network of teams and tools. By becoming more adaptive, more expressive, and more awake to change, teams will be enlivened and engaged in ways that are difficult to predict. What are the implications for solopreneurs, small teams and massive organizations? Rather than ask this as a technology question, let’s ask it as a question about culture, about how work gets organized, and about who is permitted to create. As AI researcher and author of Co-Intelligence Ethan Molick puts it:

“The future of work isn’t just about individuals adapting to AI, it’s about organizations reimagining the fundamental nature of teamwork and management structures themselves. And that’s a challenge that will require not just technological solutions, but new organizational thinking.” — Ethan Molick

The Essential Cultural Seeds

Support for personal software at organizational scale requires a more porous approach to roles and a different permission structure. This invites IT and governance functions (historically the necessary villain of the organization) to shift from gatekeeping to enablement, from enforcing uniformity to cultivating safe conditions for experimentation and perhaps most importantly for the organization as a whole to benefit from these micro-creations. This is not some fantasy or abstract ideal. It is already happening inside highly regulated, multinational institutions. As we’ll explore here in our own case study, we’ve seen a global bank create the conditions for thousands of employees to shape their own tools without setting their compliance requirements on fire. Companies and governments often use the fact that they are regulated as a kind of heat shield falsely protecting them from edgy innovations. This blanket excuse nurtures a risk of hiding from the uncertainty of innovation rather than risking value erosion by protecting its historic successes.

For those tasked with enabling innovation at scale, the ability for someone who has expertise and authority in a process to improve it opens a space that has always existed but has rarely been reachable. Talent remains one of the largest investments organizations make. Strategic dedication of one’s time remains at the core of leading a high-performance team and organization. Through this lens, inviting many to experiment with creating new tools may appear a waste of a precious resource. To be clear, we are not implying a free-for-all. Teams still must choose where to focus and benchmark outcomes.

From Permission to Practice

As teams adopt AI across the organization, relevant and timely questions of responsibility, visibility, and coherence surface quickly. These questions are a strong local signal for attention and offer opportunities to lighten permissions and reveal new opportunities for value and risk.

As the actors of a team or process are empowered to reimagine it and then to animate that with modified tools, there will be many insights, observations and surprises. Most will be difficult to ignore as they will surface process efficiencies and effective alternatives. They will also lead strong teams to confront bold questions in the context of a more fluid organizational model and toolset.

Some organizations will resist, as it can be uncomfortable. Others will recognize it as a chance to broaden who gets to participate and improve how work happens or how it could. By welcoming more people into the design process and granting permission to refine and explore, leaders are also retooling teams and creating a culture where learning is prized and strengthened.

What follows is a look at what personal software can become once it enters organizational life. These examples are intended to show how personal, human-scale tools can accumulate into real organizational learning and relevant capability. We also illustrate how a culture of permission and trust enrich the value creation engine of an organization and shape what people attempt.

New Archetypes Enter Our Offices

(hint: they’re software)

Enter into a meeting with OIO Studio, a design firm in Barcelona specializing in emerging technologies, and you will notice something new. A set of AI tools has become part of their toolkit. Some are current commercial software like Replit, a coding assistant that transforms natural language into code. Other tools are personal software created by members of the team for their often recurrent and specific needs, like a custom AI agent to synchronize their online agenda. Another type of personal software here is Roby, the team’s AI creative director. These are lightweight, highly iterative tools whose anticipated short lifecycle is outweighed by their near perfect fit for today’s tasks.

This is how the team describes Roby: “Roby is OIO’s first non-human AI creative director. He/It helps us to come up with new ideas and products, moderates the Discord community and runs its own Instagram account.” Roby first lived in a Raspberry Pi and then a Mac Mini under a desk in OIO studio. Some of the code is open source.

Roby is a good example of AI software archetypes that are entering our world through our homes and offices. Roby is not a product OIO bought. It is not software commissioned from an agency. It is a tool shaped by one studio’s practice, built by the people who use it, for purposes no vendor anticipated.

OIO operates in a space where experimentation is daily practice, not special initiative. As creative technologists, they help clients transform emerging AI capabilities into new experiences and services. Not everyone will build an AI creative director that lives under a desk. But look around, and you might recognize OIO as a weak signal of what could soon become ordinary: teams assembling their own tools from AI capabilities, shaping software to fit their specific practice.

The Long Tail of Unmet Needs

Most people we speak with — and ourselves, honestly — are constrained first by time. We can imagine long lists of problems worth solving but cannot picture finding the hours to explore, test, and refine solutions. Inside teams and across organizations, the same pattern repeats: ideas that address real inefficiencies exist, but if the company hasn’t budgeted, prioritized, and incentivized people to address them, they never make the list. Or they simply fall off it.

Over nearly two decades of digital transformation work, we’ve watched this play out in organization after organization across industries and time zones. IT departments cannot cover all needs. When the various users have their unique problems, a centralized team is incapable of serving all. IT departments that learned to survive demand by learning to say no. They standardize: vendors, tools, process and expectations. They prioritize. This is rational. But it leaves a restless long tail of niche needs unserved, quietly draining energy and other assets.

When official channels cannot help, people find other ways. Teams adopt unauthorized tools. Workarounds proliferate. Processes migrate into spreadsheets, messaging threads, and personal notebooks. This practice — the shadow IT — is typically framed as a problem to eliminate: a compliance risk, a security gap, a governance or cultural failure. But shadow IT is an important signal. And it’s one worth reading carefully.

When people route around official systems, they’re not being reckless. They’re being resourceful. They’re telling leaders, in the clearest way available to them, that a need exists that the organization has failed to address. The workaround is the message. Shadow IT reveals where the gaps are — which teams are most frustrated, which processes are tearing or already broken, which problems have been waiting for a solution that never came.

For middle managers and team leaders, this signal deserves particular attention. When someone builds an unauthorized tool or adopts a platform IT hasn’t approved, the instinct is often to shut it down. The more useful instinct is to ask: what does this tell us? What problem were they trying to solve? What does it say about our systems and our changing requirements? What haven’t we provided? Is there curiosity alive here — experimentation happening, learning accumulating — should our team and organization be cultivating these instincts rather than suppressing them?

Personal software makes these gaps visible in a form the organization can learn from. Like desire paths worn through grass in a public park — more honest and more useful than what the planners originally laid down — they show you where people actually need to go. The question is whether to follow the path or keep forcing the old map to be obeyed.

Zooming out, for organizations, personal software represents a wide opportunity to immediately tackle problems that formerly were not important enough to invest time or resources. The benefits are many beyond solving the specific issue. As tools spread and more people throughout organizations try their hand at defining, designing and building, talent grows more confident, resilient and creative.

There is also an emotional cycle that accompanies this shift. Initial excitement gives way to a sense of invincibility. Shortly after, a wave of doubt often follows: if anything can be built quickly, what is worth building at all? This arc mirrors how people respond to any profound change. Personal software functions as a safe entry point into that cycle. It allows individuals and teams to experience capability without committing to grand reinvention. In that sense, it is less a product strategy and more a learning strategy.

Not surprisingly, these broadly distributed capabilities also raise strategic questions. If every employee can build her own tools, what happens to procurement, governance, risk management, human resources, and institutional knowledge? How can an organization support this capacity without losing coherence or its ability to focus performance? How can it capture the experiments, learning and ultimate work that emerges from thousands of small experiments?

Consider a large multinational organization. Within it, a legal advisory team of nine lawyers handles 40,000 customer queries per year. They need to consolidate product specifications, internal policies, and external regulations into one searchable system. This is too specific for commercial software: the combination of sources, the particular regulations, the internal policies are unique to this organization. It is too small for the IT department to prioritize: nine people out of 120,000 is not a compelling business case. So the need persists. The lawyers spend hours on lookups that could be faster. The frustration accumulates quietly. We will return to this team later. One of them eventually built a solution.

Personal software offers a perch. From it, individuals can glimpse what new capabilities feel like without dismantling the entire operating model. It is a way to touch the future lightly. Like a vaccine, it introduces a controlled exposure rather than a sudden system shock. Teams begin to sense what is possible before being forced to reorganize around it.

Shift: From Build/Buy To Build/Buy/Enable

For decades, organizations faced a binary choice when they needed software. Build: commission custom development from internal teams or external agencies. Buy: purchase commercial software and adapt workflows to fit. Both options have constraints. Building is expensive and slow; it requires technical resources and competes for IT attention. Buying means accepting software designed for generic users, not a specific context. The long tail falls through the gap: too small to build, too specific to buy.

What we’ve been talking about here is a third possibility — Enable. Organizations who welcome this capability can now enable their employees to create their own software. Enable does not replace Build or Buy; it addresses different territory. Build and Buy continue to serve the head of the curve: large-scale systems, mission-critical infrastructure, widely shared tools. Enable serves the long tail: specific needs, small teams, local workflows, contextual problems. The question is no longer just “should we build or buy?” but “should we build, buy, or enable?”

Enable works for the long tail because the people closest to a problem understand it better than any IT department or vendor could. As Ethan Molick observes:

“Individual workers, who are keenly aware of their problems and can experiment a lot with alternate ways of solving them, are far more likely to find powerful and targeted uses of AI.” — Ethan Molick

They can iterate quickly, adjusting to fit. They do not need to justify a business case to a committee; they just need to solve their own frustration. There are other consequences here too — inviting a wider group of people to resolve issues and improve workflows and outcomes builds a greater sense of ownership for the work, team and overall performance. Not everyone will jump in, but those who have imagined altering an antiquated process or shifting permissions are likely to be delighted by this new ability to boost operations. This represents some of the cultural shifts and cues that welcome non-engineering teams into the work of enriching system efficiencies.

AI adoption at this stage becomes a portfolio decision. Core systems, data infrastructure, and mission-critical tools warrant centralized control and dedicated teams. But niche tools, team workflows, and contextual problems are different territory: this is exactly where distributed creation thrives. The question isn’t whether to centralize or democratize; it’s knowing which tier you’re operating in.

In 2024, we helped one large financial organization explore their path through this. They decided to test whether AI could help anyone, not just specialists. What happened next offers a map for those choosing to develop their capability to enable.

Global Bank Case Study: BBVA

That organization was BBVA, a Spanish multinational bank with 120,000 employees and one of the most advanced AI adoption programs in Europe.

In 2024, BBVA entered into a partnership with OpenAI to provide 3,000 ChatGPT Enterprise licenses to employees. They faced a crucial choice. Elena Alfaro, Head of Global AI Adoption at BBVA, describes it:

“Should we give everything to the data team, to engineering, or do we do something much more democratic and really test whether this could help anyone?” — Elena Alfaro, Head of Global AI Adoption at BBVA

They chose to democratize.

What followed felt like a fireworks show of beautiful little experiments. After a few months, thousands of employees were creating their own AI tools. By the end of 2024, employees had created 20,000 custom GPTs. Most were experiments. But 1,500 were used weekly across areas like procurement, compliance, and communication. Each serves just a few employees. Together, they form a long tail of software. By mid-2025, BBVA had expanded to 11,000 licenses. In late 2025, they sealed an agreement to provide ChatGPT to all 120,000 employees.

Remember the nine lawyers handling 40,000 queries per year? One member of that team built a custom GPT to help the group. The tool consolidates product specifications, internal policies, and external regulations into one searchable system. It drafts answers faster and more thoroughly than manual lookup; all responses undergo human review before reaching branch managers. They call it their “tenth team member.” This is software shaped by one team’s experience and workflow, not by the IT department.

The success of BBVA’s AI adoption is not one application deployed to thousands. It is thousands of applications, each serving a few. The collective impact: BBVA reports that each of the 11,000 employees with licenses saved on average three hours per week. Elena Alfaro summarizes the effect: “The clearest impact is less time. But the next is higher quality in the results, and the third is more innovation.”

Here is where it becomes interesting. Personal software has reached organizational scale while maintaining its human-scale characteristics. Each tool is still built for immediacy, still finished when it fits, still human in scale. But multiplied across 11,000 employees, the effect is transformational. This is not commercial software logic, where one product serves many users. It is personal software logic, where many products each serve a few, operating at enterprise scale. The learning here is not only incremental improvements in all corners of the organization, but a call to action for every member of the team throughout the business to observe, think and contribute in new ways. This also subtly values team and individual performance beyond the scope of their historic role as the culture of the company expands to invite anyone to shape their work. Both in theory and in practice, this shift creates a kind of talent magnet as team members have acquired essential skills and gained the ability to explore new questions and modify their tools.

What BBVA demonstrates is that LLMs have dramatically lowered the software creation barrier. Most of these tools are built through natural language configuration rather than code. The value is created when employees experiment, learn, create, and collaborate. The impact is hard to track with regular indicators and remains largely invisible in many organizations. BBVA chose to embrace these personal software assets and the learning they produce. They offer a model for others facing the same crossroads.

What Made It Work at BBVA

What is emerging at BBVA is not just a technology rollout. It is a cultural shift. Employees are becoming makers, not just users. The bank’s internal “GPT Store” lets staff share and adapt tools made by peers, spreading human-scale innovation across a vast organization. But tools alone do not create this shift. The conditions do. Three conditions made it possible.

Permission to experiment

BBVA handed out licenses and training with one clear instruction: try things. Put another way: Use it, use it, use it. The response surprised even those who initiated the program. “People were completely on fire,” Elena Alfaro recalls. “They made personal things because we had told them so clearly to experiment.” This “on fire” behavior shows that personal software flourishes when organizations grant cultural permission, not just technical access. It is not enough to provide tools. They must signal that experimentation is welcome.

Trust and autonomy

Distributed experimentation requires trust. BBVA gave employees autonomy with responsibility. They did not create a “transformation office” that dictates what employees can and cannot do. Instead, they created space for employees to solve their own problems, with guardrails that emerge over time. Trust is the foundation. Without it, employees wait for instructions rather than act on frustrations.

Community over code

Personal software creators share knowledge, not code. This is different from open source, where the code itself circulates. At BBVA, people share how they solved problems, what prompts worked, what failed. A community of practice formed around this exchange. A GPT created in Mexico may inspire a different solution in Turkey. When asked about duplication, Elena offers a surprising perspective: “How do we prevent people from repeating work? I tell them this is the least of our problems. If two people build two GPTs, they have both learned to create applications, which is far more valuable. And they probably bring complementary perspectives.” Unlike the traditional approach to software, duplication is not a waste. It is learning.

Underlying all of this is a demystified view of AI. Elena describes it simply: “It’s a tool. You need to know where it’s useful and where it’s not.” This practical, problem-first approach is the same attitude that defines personal software. No hype, no innovation theater. Just people solving problems with new capabilities.

Culture creates the conditions. Governance sustains them. Experimentation without structure eventually produces chaos. The next question: how do you govern thousands of personal tools without killing the energy that created them? Don’t ask how fast we can automate, but how deliberately we can preserve the human creativity that makes adaptation possible in the first place.

And, underlying that, how might the culture that makes good work possible be expressed directly in the systems we empower people to build?

What Failure Teaches

Not every experiment at BBVA worked. Of 20,000 GPTs created, most faded. Some answered questions incorrectly. Some duplicated tools that already existed. A few were built with enthusiasm and never opened again. This is not a cautionary tale. It is the point.

Personal software at organizational scale is, by design, a learning engine. The value is most definitely not that every tool attempted succeeds. It’s in the process of building, testing, failing, and revising and what that teaches the people doing it. Elena Alfaro’s observation about duplication captures this well: two people building two GPTs that solve the same problem haven’t wasted effort — they’ve both learned to build software, and they probably arrived at complementary insights potentially via alternative routes. Learning is the product, not the tool.

This requires a particular tolerance — not just institutional, but personal. The emotional arc of building with new tools tends to follow a recognizable shape. Early experiments feel exhilarating. Then something breaks, or doesn’t work as expected, or turns out to be less useful than imagined. A wave of doubt follows: if anything can be built quickly, what is worth building at all? Caution! This is the moment many organizations inadvertently kill — by measuring too early, by pulling governance levers prematurely, by treating the pause as evidence that the whole initiative was misguided. The organizations that navigate this well share a common posture: they treat early failure as information rather than verdict. Some practical patterns we’ve observed: When a tool doesn’t work, ask why before asking who. A failed experiment usually reveals something about the problem’s framing, not the person’s judgment.

At present, AI systems assist. They generate drafts, surface options, and suggest structures. Determining whether a problem is truly solved still rests with the human who lives inside it. That judgment cannot be outsourced. Yet the path to that judgment is changing. In the next chapter of this series, where we’ll focus on the individual, we’ll explore the notion of “mechanical sympathy” — the ability we have, which goes beyond curiosity, to develop new capacity as we learn from how things break when we just try. Rather than requiring formal technical training, competence increasingly develops through iteration. A tool is built. It almost works. It reveals its gaps. It improves. No disasters happen. Through that cycle, users acquire an embodied sense of what these systems can and cannot do. Experience becomes literacy.

When adoption drops off, treat it as data. A tool that fades after two weeks tells you the problem it addressed was either solved or wasn’t actually as pressing as it seemed. Both outcomes are useful. When duplication happens, celebrate it rather than managing it away. Parallel experiments surface different approaches to the same problem. Convergence happens naturally. Consider: premature standardization prevents discovery. When something goes wrong with real consequences — a tool that surfaces incorrect information, a GPT that miscommunicates policy to a colleague — treat it as a governance signal, not a reason to slam the aperture closed.

These moments are precisely when the tiered governance approach matters: not all tools require the same level of oversight. Some need strong guardrails and failure helps teach which ones those are. The deepest failure mode isn’t building something that doesn’t work. It’s building something, watching it fail, and concluding that the whole endeavor was not valuable or altogether too risky. Classically, the signal that ‘this isn’t working’ tends to arrive just before the real learning begins. Hang in!

The Honda Approach

Honda offers a useful counterpoint here, because the story starts with culture rather than tooling. For decades, the company’s core philosophy of Waigaya has served as a mechanism for frank, high-energy debate across domains. It exists to surface assumptions, force contact between different kinds of expertise, and keep ideas moving through friction rather than ceremony. That practice matters in a company whose work requires tradeoffs among safety, performance, manufacturing realities, regulations, and design.

In 2025, Honda’s internal AI group tried something unusually direct: they treated that cultural practice as a design pattern for AI agents. By articulating the culture as a template, the move to create agent software became straightforward. A Honda team built a multi-agent system in which different LLM-based agents act with various domain expertise and perspectives. They then engage and coordinate through discussion, reflecting how the Honda culture works in practice. Central to this is the idea that experts gather and openly discuss and debate their way toward a solution. The team tested multiple agent discussion styles — decentralized, centralized, layered, and shared pool — and after experimentation found the decentralized approach to integrate diverse views more naturally. In other words, they asked their AI agents to argue the way we’ve enjoyed watching Honda engineers and designers engage and respectfully argue. That decentralized style bias was designed into the system with the Waigaya cultural practice in mind.

What Honda did was not simply add AI to its culture. It studied its culture and asked how to best represent it computationally. This is a rather significant shift in approach. It begins with a commitment to preserve the culture while modernizing the tools and teams. It suggests a path into AI adoption that begins with cultural intent. Rather than handing out licenses and hoping for good outcomes, an organization benefits from first asking what kinds of engagement it values and wants to protect: how disagreement is handled, how different forms of expertise meet, how decisions converge or diverge, how assumptions are challenged. Answering those questions moves learning from invisible to foreground — it becomes a central design element, not an afterthought. Honda’s experiment treats these as first-order requirements. Their system design assumes that the form of conversation shapes the quality of outcomes, then operationalizes and tests that assumption in practice.

This fits the broader pattern we are seeing across the chapter. BBVA shows what happens when an institution changes permission structures and creates room for distributed making. Honda’s Waigaya work shows another dimension: some of the most important work involves strengthening, protecting, and even encoding the conditions under which good judgment emerges. If we want to move into this world in a culture-first way, this is part of what that can look like: an organization that studies its own generative practices and then builds systems that reinforce them.

What Happens When This Spreads

The tools multiply. So do the implications.

Governance becomes an innovation tool

Governance means more than a one-time imposition of rules. Culture creates energy. Governance must sustain it without extinguishing it, by cordoning off spaces for expanded experimentation. Traditional IT governance assumes centralized control, standardized processes and approval workflows. Personal software at scale breaks these assumptions. You cannot govern 20,000 experiments the same way you govern a traditional large-scale system. A new model is needed. When designed properly, governance can be the means by which to observe and learn, and continuously update models and best practices as they evolve at speed.

Elena Alfaro poses the question directly: “How do we govern all of this? Can we govern everything?” Her answer: “No. You have to let people create their software, then govern what is heavily used or addresses a relevant process.” In the first seven months, employees launched more than 3,000 GPTs, of which over 900 were flagged as cases of strategic interest and with upscaling potential. A few months later, 1,500 GPTs were in active weekly use. Of those, only a fraction fall under strict governance. This is a radical departure from traditional governance. Most personal software lives ungoverned, and that is fine. It serves a few people, solves a small problem, and fades when no longer needed.

For the 1,500 tools that cross the threshold, governance is lightweight but real. Ownership: who is responsible for this tool? Data curation: what data feeds it, and is that data appropriate? System prompt quality: is the tool well-designed and safe? Impact measurement: is it actually helping? These checks create accountability without bureaucracy.

For the tools that touch relevant processes or serve many users, oversight is stricter. Compliance review, security checks, integration standards. The distinction is not arbitrary. It emerges from use patterns. Governance follows adoption, rather than preceding it.

New roles emerge

This model requires new roles. At BBVA, creators are called “Wizards”: employees who build personal software. Champions hold an overall vision for a domain and connect efforts. In other organizations, Supervisors review tools that reach the governance threshold. Curators monitor data quality and model drift. These roles do not replace IT. They extend the organization’s capacity to manage distributed creation.

New risks emerge

The entry barrier is low. This is good news and bad news. When 20–30 percent of employees can create software, up from 2–3 percent, the potential for mistakes grows. Model drift means AI behavior changes over time; tools may degrade without anyone noticing. Dependence on specific LLM providers creates strategic risk. Personal software may access sensitive data without proper controls. The risk is not that employees will create. It is that creation will happen without visibility (Shadow AI).

Measuring success

Traditional software metrics do not capture the value of personal software. Uptime, adoption rates, cost per user: these measure products built for scale. BBVA tracks time saved, quality improvement, innovation capacity. These are closer to HR metrics than engineering metrics. The question shifts from “how many people use this tool?” to “did this tool help these specific people?”

BBVA’s model is not the only way, but it offers a template. Experiment first, govern later. Classify by impact. Identify and measure what matters. This approach worked, in part, because custom GPTs always keep a human in the loop. As personal agents — tools that act, not just assist — begin to enter the workplace, governance will need to evolve. The question is how to keep control and avoid risks without sacrificing the agility that made the experiment worth running.

For teams and organizations, the work starts with culture. Some questions to ask:

• How ready is your team to try on new ideas?
• What is missing to unleash cheap, short experiments for roles who have only been on the receiving end of new tools and systems?
• How can you harness insights and hunches that are floating around without a home?
• Whose job is it to design better practices when the work keeps changing?

And, perhaps most importantly, where would you begin to make this real?

If you have seen your organization in these pages — whether at the crossroads or already experimenting — we want to hear your questions, your experiences, and the examples you are seeing in your own context. Reach out on LinkedIn or join us at Pre-Alpha Club.

To be notified when the next chapter is published, follow us on Medium or LinkedIn.

This work is part of the Technology Foresight program of armasuisse, the Swiss Federal Office for Defense Procurement. It would not exist without Quentin Ladetto, his vision, his questions, and his insistence that this kind of work be made public. Foresight shared is a path to greater strength and resilience. It is a reminder that software, and thinking about the future of software, is everybody’s business particularly in the domain of defense and security.

Thanks to Patrick Pittman for his numerous contributions to this essay.

In 2019, a friend reached out with a simple request. As a start-up founder and an avid cyclist, he needed a simple tool to share bike ride summaries on social media. Nothing fancy, just a clean, shareable map animation. As the experiment worked, I decided to make the tool accessible to anybody with a similar need. It worked. Other cyclists saw it, and then hikers, navigators, runners, anybody who records GPS trails wanted to create their own map animation. That quick solution has become Rumbo, with over 80,000 videos produced by thousands of users and counting.

I didn’t start Rumbo to “revolutionize” or “disrupt” anything. I certainly did not plan to build a commercial product. I know how to build software, but I never fall into the distraction that every software idea must scale. I just solved a specific problem for a friend. And yet, here we are, somewhere in that uncomfortable zone between a personal tool and a commercial product with more than 500 Pro customers, raising questions I never intended to face about maintenance, scaling, and sustainability.

Rumbo is part of what I call “personal software”, a fourth category of software that sits alongside commercial software (built for scale), boutique software (tailor-made for companies and organizations) and open source software (structured for collaboration). Personal software is a solution that is good enough for a niche audience. It’s software made by “anyone,” for themselves and the people around them. This is fundamentally different from software designed by professionals organized in teams for paying customers.

The schema above illustrates the four main software categories that differ not only in what they produce, but in how makers and users relate. Personal software keeps everyone in one circle. Boutique and commercial software place a company between contributors and users. One collaborates with a few, the other serves many. Open source distributes contributors and users roles across a decentralized community.

For decades, software has been something most people use, not create

Creating software has been the domain of trained professionals like me. AI is changing that. Today, it is becoming more accessible to anyone who is “programming-adjacent” (e.g. an academic who builds their own tools, a creative who experiments with code, an expert who wants to try building a solution). Soon, more of us will have the possibility to build our own micro-tools, not as programmers, but as people with ideas or problems to solve.

This essay maps that emerging terrain: what personal software is, how it becomes popular, what is changing, why it matters, and what motivates its makers. It is also an invitation to an attitude and practice to rethink software that is beginning to form with AI.

What is personal software

Personal software is not entirely new. People have been building their own solutions for decades. Some of my favourite examples come from the early Mac indie scene, like NetNewsWire, an RSS feed reader I still use today that Brent Simmons originally built in 2003 for himself and a few friends. This ethos has appeared under many names over the past decades.

In 1973, at the dawn of the Personal Computer era, philosopher Ivan Illich introduced the idea of tools for conviviality, which empower individuals to shape their own technological environment with autonomy and creativity. Later in 2004 during the mobile and social media revolutions, Clay Shirky reflected on his students work at NYU’s Interactive Telecommunications Program (ITP) and described what they were building as situated software designed for particular social contexts. Today, people talk about single-player software, disposable software, malleable software, moldable development, personal micro-software, or one of my favorites snowflake software. The terminology varies, but the essence is the same: there’s a type of software that does not fit neatly into the well-known categories.

I prefer “personal” because it describes what the software creates, not just what it is. It emerges from a personal relationship (a friend’s request, a colleague’s frustration) or it establishes one. Each Pro subscriber of Rumbo left me a message explaining why they signed up. The app is part of a relationship, not a transaction.

The key features of personal software

According to my experience, observations and research, what distinguishes personal software from its commercial, tailor-made and open source cousins are 3 key characteristics:

1. Built for immediacy

Personal software begins with an immediate need, a problem faced by its creator or someone nearby (e.g. friend, colleague). It’s not designed for a market, but for a moment. It solves something specific without trying to generalize or anticipate future use. If others find it useful, it can accidentally break toward commercial territory, but that’s a side effect, not the goal.

2. Finished when it fits

Personal software is complete the moment it works for its maker. It doesn’t need to scale, integrate, or evolve into a product. It’s not a Minimum Viable Product (MVP) waiting for investors, nor a demo that showcases technical capacity, nor a Proof of Concept (POC) waiting for validation. It is done when it fits. Longevity depends on continued use, not planned maintenance or future versions.

3. Human in scale

Personal software is designed for use by a specific group, rather than for a generic set of “users”. It optimizes for specificity over universality, intimacy over scale, immediacy over evolutivity. Its makers learn by building, staying close to the code and the group of people who use it. The learning from the creation and adoption (not the code) can inspire, influence or guide the development of other categories of software.

What makes personal software distinctive is proximity: it is made by people who understand the problem more deeply than the underlying technology to solve it. For this reason alone, broadening who can make software is essential.

How personal software became possible

In 1972, when computers were “mainframes” usable by a few individuals, computer scientist Alan Kay imagined personal computers as tools that ordinary people could program themselves. Since then, along with the Personal Computer era, many approaches have sought to democratize software creation: simpler languages, WYSIWYG systems, mashups, and macro tools to name a few. The mobile revolution got us a step closer to make Alan Kay’s vision a reality. Although impossible to measure precisely, tens of thousands of apps on the App Store are built by one or two individuals and mostly for a tiny audience.

For instance, novelist Robin Sloan built BoopSnoop, a messaging app for his family of four. It captures photos and videos, sends them without editing, and messages disappear once viewed. For years, it has maintained four daily active users with zero churn, a resounding success by its creator’s own measure. Sloan’s essay about the project coined the term “home-cooked software”: software made like a meal, for the people you love, with no need to scale.

I experienced the challenge of democratization firsthand in 2012 with Quadrigram, a visual environment built within the design studio Bestiario to let people manipulate and visualize data without writing code. While it opened new possibilities for non-technical professionals like designers, analysts and journalists, it also revealed a limit: the promise of software made by “anyone” is not really for everyone. Even with highly visual tools, creating software still requires skills in abstraction that most people do not naturally possess. The vision of software made by “anyone” remains powerful, but it is not evenly accessible.

What we did not imagine is that 10 years later, people would be able to produce similar data visualizations through conversations with machines.

Why personal software is gaining momentum

In 2023, OpenAI’s co-founders Andrej Karpathy famously claimed: “The hottest new programming language is English.” It marked a transformative moment in software development where Large Language Models (LLMs, including multi-modal, visual models) could effectively link natural language with functional code.

LLMs serve as authors generating code. AI coding assistants and agentic coding based on LLMs (Cursor, Floot, Claude Code, Replit, OpenAI’s Agent Builder) leverage a wide amount of code repositories and web conversations to generate code from natural language. They also use both open source and private libraries and frameworks to automate software production. Some call this evolution the rise of industrial software.

The emergence of LLMs has rapidly changed the software landscape. In 2025, developer surveys indicate that nearly all professional programmers have experimented with AI coding tools, with a large majority using them weekly.

That rapid adoption of AI is made possible by human collaboration

Today, there is a large amount of online tutorials, forum conversations and YouTube videos teaching non-professionals how to create software with LLMs. For instance, inspired by OpenAI demos at the release of GPT5, the tech popularizer Azeem Azhar described in his Exponential View newsletter how his team without specialized knowledge had built a Korean learning app. Similarly, tech journalist Kevin Roose, who does not code, built a replacement for his favorite read-later app in under three hours. This pattern repeats across social media. Designers, product managers, academics sharing their small apps and simple tools that would have needed a team of developers a few years ago. Together, they explore a wide terrain of new possibilities demonstrating how to create software without formal training. Similar developments are taking place in organizations that I describe in more detail in a follow-up chapter.

The main personal software archetypes

We are clearly in the early stages of a significant new wave of capabilities and reach. Looking at what people share and talk about, few archetypes emerge from traditional code to LLM-infused systems:

In some cases, AI models (including LLMs) are components embedded within applications, wrapped by code that provides structure, guardrails, and interfaces. People integrate these AI models into their solutions using a mix of traditional code, visual programming and natural language (i.e. prompts).

Examples

An assistant that saves one team ten hours a week. A workflow that simplifies the work of fifty people. An app that brings a family together. A plugin that augments your creativity. Stories like these circulate online every day. Here are a few from my own circle:

Tsundoku
Archetype: Plugin
At Próximo Lab, we were frustrated by the lack of social tools to share interesting articles among ourselves. We built Tsundoku, a Chrome extension that remembers what you read and links it to everybody’s readings. It surfaces contradictions, complementary ideas, and connections that enrich our conversation. It is tailor-made software by a small cohort for itself.

Shopify-OML
Archetype: Plugin
Próximo Lab member Rohit Gupta coined the term “in-between software” for the half-baked mechanisms that connect services never designed to work together. Our editor Patrick Pittman built one: a Shopify plugin linking our online bookshop with a local distributor who ran warehouse stock on a Google spreadsheet. Personal software often lives in these gaps: too specific for any vendor to address, too necessary to leave unfilled.

Tenth team member
Archetype: Assistant
At BBVA, where I have consulted on AI strategy, a member of the legal advisory team built a custom GPT to help the nine lawyers handle 40,000 customer queries per year. The tool drafts answers faster and more thoroughly than manual lookup; all responses undergo human review before reaching branch managers. It is software shaped by one team’s experience and workflow, not by the IT department.

Flow
Archetype: App/Workflow
Designer and teacher Etienne Mineur has been creating custom tools that augment a designer’s singularity, instead of AI normalizing creation. He built Flow, a workflow generator that orchestrates a synthetic creative team (strategist, logo specialist, design critic) to brainstorm, critique, and iterate under his direction. It is software shaped by one person’s practice and AI’s possibilities, not by a market.

What personal software creates

Unlike traditional business models (e.g. enterprise licensing, contracts for tailor-made development, paid support and hosting for open-source), personal software rarely generates direct revenue beyond to cover costs. Its value lies elsewhere.

People empowerment

When existing tools are too generic, too expensive, or too complex, or when no tool exists at all, people look for ways to build their own. Personal software emerges from this impulse to shape one’s tools: fixing frustrations, automating tedium, augmenting capabilities, creating something new within a specific practice.

In a world defined by software built for a generic user, there is something powerful about creating tools to meet one’s own needs. Following the spirit of the Do-it-yourself (DIY) movement, personal software may lack the consistency of professional products, but it develops a different quality: the charm of something made by hand, for oneself, friends, family and colleagues. Personal software is built by and for real people with names and contexts. As Robin Sloan demonstrated with his family app BoopSnoop, software can be meaningful and valuable without pursuing commercial success. The measure of success becomes people empowerment, not financial growth.

Knowledge creation

Because personal software does not aim at robust, perfect, or complete results, the code itself has little transferable value. It does not cover edge cases and it may only work in a specific context. For these reasons, It rarely converts directly into open-source projects or commercial products.

The real value emerges when people experiment, learn, and share not the code, but the knowledge. Creators of personal software describe how they solve problems, what they learn from a small set of users, what works and what does not, what they imagine building. This knowledge and creativity circulates more through language (e.g. conversations, online posts, and video demonstrations) than code repositories. The impact is difficult to track with conventional indicators and remains largely invisible.

It is through cycles of learning, collaboration, and refinement that some personal software move toward other software categories. For instance, NetNewsWire traces a compelling arc. In 2002, Brent Simmons released it as an RSS reader he built for himself and fellow Mac users who wanted to follow blogs. It was personal software in its purest form. The app found an audience. In 2005, NewsGator acquired it, and Simmons joined as an employee, continuing as lead developer while the product moved into commercial territory. Six years later, Black Pixel acquired it; Simmons no longer worked on the code. The app drifted. Then in 2018, Black Pixel returned the name to Simmons, who merged it with Evergreen, an open-source reader he had been building independently, and released NetNewsWire 5.0 as free, open-source software. The application traveled from personal to commercial to open source. Today it remains what it was in 2002: software made by someone who uses it, with and for others who share the same need.

Mutual enrichment

The incompleteness of personal software, far from being a flaw, becomes an invitation. In his 1989 essay Worse is Better, computer scientist Richard P. Gabriel observed that systems designed with practical simplicity often achieve better adoption than perfect feature-rich systems. Imperfection creates openings: releasing something that others know how to improve is one of the best ways to attract collaboration.

Personal software is the beginning of a relationship, not the end of a transaction

The media and software company Every has built part of its business model around this principle. They offer five AI tools each produced by one person: Lex for writing, Spiral for content repurposing, Sparkle for file organization, Cora for email, and Monologue for voice dictation. These products first emerged from personal needs of the 11 employees, then found resonance with their audience. They are bundled into a $200/year subscription alongside Every’s in-depth writing on AI. They are not as polished as commercial software from a dedicated product company. Rather than building products for growth and generic users, Every builds products for itself and within its community of subscribers.

This inverts the conventional relationship between software and users. Commercial software extracts value from a user base. Personal software creates value within a group. The distinction matters: one scales by acquiring customers, the other strengthens by cultivating relations.

Software as an invitation to the future

This essay has mapped a terrain that is transforming. We are standing in what serial entrepreneur Lisa Gansky calls the “No More / Not Yet”: the traditional models are challenged, but the alternatives remain unfinished. Personal software is an invitation to shape what comes next.

For decades, software has been something most people use, not create. The software industry measures success through scale: users, revenue, growth. That relentless pursuit of scale has produced systems that serve their own logic rather than the people using them. The tools we use daily are shaped by distant teams optimizing for markets we happen to belong to. Personal software inverts this relationship and the measures of success: it is built by the people it serves, small enough to comprehend, close enough to control, good enough to fit.

The barrier to software creation has lowered, but it has not disappeared: even with AI assistance, systems still demand comfort with abstraction. LLMs are expanding the circle of who can build but they do not erase the circle entirely. Projects are easier to start, but making them fit still requires ability and expertise. In the next chapter, I look at the new creators that shape the practice: their approach, their work, their skills.

For organizations, personal software raises strategic questions. When employees can build their own tools, what happens to governance, human resources, procurement, and institutional knowledge? How do you support this capacity without losing coherence? How do you capture the learning that emerges from small experiments? How do you measure success? I explore these questions in the final chapter.

Rumbo started with a friend, a problem, and a few days of work. Many well-known software companies and other commercial businesses start that way (e.g. eBay, Slack, BlaBlaCar, Wallapop, etc). The difference now is that more people and organizations can shape their own tools with AI, forming a long tail of thousands of micro-tools.

If you have built personal software, or if you are beginning to experiment, I want to hear your story, your approach, and what you are learning along the way. Reach out on LinkedIn.

To be notified when the next chapter is published, follow me on Medium or LinkedIn.

This work is part of the Technology Foresight program of armasuisse, the Swiss Federal Office for Defense Procurement. It would not exist without Quentin Ladetto, his vision, his questions, and his insistence that this kind of work be made public. Foresight shared is a path to greater strength and resilience. It is a reminder that software, and thinking about the future of software, is everybody’s business particularly in the domain of defense and security.

Thanks to Eva Fernández García, Lisa Gansky, Matthieu Gioani, Rohit Gupta and Andrés Ortiz for their thoughtful suggestions on this essay.

Software Gets Personal: An Introduction was originally published in Próximo Presents on Medium, where people are continuing the conversation by highlighting and responding to this story.

I come from an engineering and research background, where structure, precision and clarity matter above all. But I have always been drawn to great storytellers (e.g. novelists, history popularizers, comedians, science communicators). There’s something almost magical about their ability to weave ideas into compelling narratives. My own writing has always been more pragmatic: detailed specifications, internal notes synthesizing multiple perspectives, documents laying out a vision, research papers sharing results of experiments. Over the last twenty years, I learned that the person who writes things down holds a particular kind of power. The power to shape how ideas spread and take root both in my own mind and with others when I share them.

For me, writing has never been comfortable because I write to think. Writing forces my mind to meet reality: to capture multiple points of view, to form an opinion, to articulate a perspective. Without writing, my ideas remain vague, badly formed, and poorly sustained.

I often share short texts (like this one) with a colleague for a quick review and sometimes I link them to publish an essay, to structure a class, or to give a public presentation. But mostly, I accumulate stacks of half-baked drafts that serve as notes for ongoing thinking. Until recently, this was largely a solitary struggle. It still is at its core. The thinking, the choosing, the crafting of ideas remain mine alone. But the surface of writing, the articulation itself, has changed.

Writing becomes conversational

With the arrival of Generative AI tools, some parts of my writing process take place via a messaging interface. Some call it “vibe writing”. Suddenly, I could engage in a dialogue about my ideas, testing different ways to articulate them, and helping me when I get stuck. For someone who struggles with storytelling, this felt revolutionary. I had a tool to help translate a messy train of thought in my head into fluid, readable texts, almost like a storyteller.

I lean on the popular general-purpose LLMs (e.g. ChatGPT and Claude) to polish the flow of my thoughts and AI wrappers (e.g. Perplexity) to find references and novel research material. I use these tools to integrate my notes, find the boundaries of my text and to better position my thoughts. They prove particularly useful to prevent me from “Magnum Opusing”: that familiar trap where the scope expands endlessly, and notes accumulate until the project becomes too complex to ever complete.

If I had to illustrate my writing trajectory for this text, it would resemble a wandering line slowly gaining definition. In the diagram above, the blue line marks how the text evolved as a living document through successive revisions, moving from vague fragments toward clearer, more mature thoughts. The light grey lines represent notes from observations, readings, and conversations that intersect with AI-assisted articulations in pink. Together, they form an exploratory path that splits into multiple tracks, loops back, and eventually converges. What begins with a narrow scope widens into not just a finished text, but a broader landscape of thinking that only emerges through the act of writing.

This diagram mirrors my other creative work in software engineering and futures design, where Generative AI tools have amplified my capabilities. I have been vibe coding ideas and prototyping concepts without deep knowledge of the latest software libraries or fluency in certain programming languages.

Having spent years creating software and envisioning futures, I have learned to watch carefully for what we might lose in what we gain. In 1964, Marshall McLuhan argued in Understanding Media, that technology and society co-evolve: every augmentation is also an amputation.

The hidden costs of writing with AI

LLMs have undeniably lowered the barriers to writing. Many people can now share their thoughts with more confidence, even when writing in a foreign language (like me now). Something that would have been unimaginable just a few years ago. We worry less about grammar and form as these feel “easily” fixable by an AI writing tool. Or is it an illusion of confidence?

Loss of authenticity

My experience of outsourcing writing often leads to safe, predictable prose; instead of pushing the limits of my original thought. If I am not careful, my ideas fade and become average. They are stripped of personal nuances, they lose their wabi-sabi and soon are no longer authentic. When approached superficially, writing with AI can easily regress thought to a statistical middle ground.

Flawed reasoning

Generative AI tools rush toward conclusions and good enough solutions, sometimes introducing illogical shortcuts. They produce plausible, compelling, confident prose that can camouflage flawed reasoning. These biases are features of the stochastic probabilistic models used in Generative AI. They are not a bug that will go away soon. As a result, I get caught in the constant effort to reappropriate the ideas, to make them my own again.

Shallow thinking

Finally, thinking is not linear as a conversation via a messaging interface. It is often erratic and messy. To grow stronger, ideas require exposure to different perspectives and the discipline of battling through them. They need time to mature. The quick-fix nature of current Generative AI tools is undeniably useful, yet it also reflects today’s shortcut culture, tempting us to bypass the kind of painful intellectual work that produces truly original and deep thoughts.

If I had to illustrate my experience of AI-only writing, the diagram would tell this story:

The trajectory jumps quickly toward apparent clarity and an “illusion of maturity”, as AI tools help me smooth sentences, tighten structure, and define boundaries. Yet the scope of the writing corpus remains relatively narrow. The final text appears complete and mature, but represents what I lose: the deeper, more erratic thinking that comes from wrestling with ideas outside the context of the interaction with a probabilistic machine.

Beyond my work, I have noticed the “illusion of maturity” in some students’ final projects and in the AI-generated content shared on LinkedIn, Medium and in blogs. Some recent studies seem to confirm these observations (e.g. Your brain on ChatGPT) though the research remains early and findings should be interpreted cautiously. The underlying driver appears to be our natural human preference for fast thinking, which is easier and requires fewer resources..

This raises a crucial design challenge: How do we keep the slow, difficult work of thinking alive when Generative AI tools make writing so quick and easy?

Writing is social

To write this text, I wanted to counterbalance my use of Generative AI tools with more human frictions. I ran a small experiment in the format of a tertulia. A concept originally from Spain, a tertulia describes a regular social gathering of people to share their recent creations and talk about current affairs. These gatherings, also called cénacle in France or salon in the English-speaking world, have long been spaces where ideas develop through human connection.

Practically, once a week, I brought together 4–6 colleagues (known as tertulianos) for a 1-hour online discussion. Each of us brought something in progress, a draft, a project, an outline for a presentation, readings, etc. We all use AI tools regularly, but the tertulia became a place to reflect and to let our ideas mature outside the rush of work.

The sessions felt like having a “writing circle” that only popular storytellers or comedians have the luxury to have. We bounced ideas back and forth, challenged each other, and offered perspectives none of us would have reached alone or with a tool. My role was to encourage these frictions, to provoke collisions, to keep the conversation difficult enough to spark new thinking. It was erratic. It was fun.

After each session, I found myself immersed in notes from our conversations, along with related observations and readings. Each session challenged and deepened my thinking about “writing to think.” I was struggling to write this very text, and that struggle felt right. It pushed my ideas past the “illusion of maturity” in ways that happened entirely outside of AI tools.

The heart of the writing process, the unexpected insights, the feeling of getting lost, the obscure cultural references, the inspiring analogies, the real leaps of imagination emerged from social interaction. Laurent articulated the constant struggle to reappropriate ideas from AI tools. Andrés introduced the concept of ‘shortcut culture’ developed by Carolina Sanín. Lisa drew connections to the book Thinking, Fast and Slow. My notes overflow with these breakthroughs, each one sparking new connections in my own mind.

A space to think

This text emerged from exactly the practice it describes: conversations with machines for focus and clarity, conversations with humans for maturity and depth. The design objective is not to replace the struggle of writing, but to make it more fertile. It points toward a larger question about how we collectively navigate our relationship with AI tools.

One quote captures well this current co-evolution between machines and humans:

“The rapid spread of AI adoption is made possible through human collaboration.”

It comes from my friend and accomplice Lisa Gansky, a member of the tertulia. Serial entrepreneur and author of The Mesh, Lisa knows what she is talking about. She is an expert on technology, collaboration and networks. Together, we share a concern: when speed replaces depth, something is lost in what makes us humans both as professionals and as citizens.

Deep and authentic thinking demands time, curiosity, vulnerability, and willingness to sit with questions. As Lisa often says, it is not a “spectator sport.” Thinking requires active making (e.g. writing, sketching, prototyping) to develop and practice the skills at its core.

The tertulia experiment shows that anyone could benefit from it. But today’s frenetic AI world needs more than occasional experiments. It needs a community of practice for this kind of slow and deliberate thinking. This is exactly what Lisa and I are pursuing at Próximo Lab: a space where a community of perpetual learners from diverse backgrounds immerse and engage with each other through tertulias and other collaborative explorations (e.g. hands-on studios, guest talks, etc). We see this type of trusted, diverse learning lab as a foundational element for our lives.

Many thanks to Laurent Bolli, Eva Fernández García, Lisa Gansky, Daniel Goddemeyer, Rohit Gupta, Vytas Jankauskas, Andrés Ortiz, and Simone Rebaudengo. This text is a byproduct of our conversations. Special thanks to Andrés Ortiz for challenging me and helping me sketch a factual diagram of my new writing practice. In memory of Nicolas Nova, with whom I would have loved to discuss all this.

Il y a 20 ans, fraîchement diplômé d’école d’ingénieur, je débarquais dans un bureau à l’Université de Genève pour débuter mon service civil. Dans l’équipe de recherche en technologies éducatives, je m’adresse à un étudiant-assistant pour ouvrir mon compte email. En le voyant bredouiller des commandes UNIX, je ne me doutais pas à ce moment-là, que cette rencontre allait fondamentalement influencer la personne que je suis aujourd’hui.

Il y a quelques jours mon ami, collègue et complice, Nicolas Nova, s’en est allé brusquement alors qu’il pratiquait ce qu’il aimait le plus: rencontrer, explorer, observer, documenter.

Sur la page mise en place à sa mémoire https://nicolasnova.org les messages laissés par beaucoup reflètent le riche réseau que Nicolas a tissé au fil du temps. Ces centaines d’hommages venus des quatre coins du monde trahissent l’évidence d’un impact qu’il préférait dissimuler sous le voile de son humilité.

Il y a beaucoup de points d’entrée au monde de Nicolas. Par exemple, certains connaissent Nicolas à travers son talent d’éditorialiste de la conférence Lift, qu’il avait co-créée à Genève avec Laurent Haug, un autre fidèle complice. D’autres se rappellent de CatchBob!, un des premiers jeux géolocalisés que nous avions développé dans le cadre de son doctorat en informatique à l’EPFL. D’autres étaient des lecteurs de son blog Pasta & Vinegar. Et encore d’autres ont été à jamais inspirés par ses observations de nouveaux gestes du numérique dans Curious Rituals, qui sera à l’origine de son doctorat en sociologie à l’Université de Genève.( Nous aimions l’appeler Docteur Docteur Nova pour chatouiller son humilité).

Au fil du temps, Nicolas avait réussi à consacrer plus de temps à sa passion pour l’écriture. Auteur prolifique, (Comment fait-il ? me demandait-on souvent) ses livres explorent des thèmes originaux liés à la culture numérique et à l’artificialisation de notre environnement. De 8-bit reggae au Bestiaire de l’anthropocène en passant par Fragments d’une montagne, il documentait ses observations de l’être humain et des technologies pour faire apparaître les nuances à contre-courant des récits simplificateurs dominants.

Nicolas était curieux de profession. Tous ses travaux sont une invitation à devenir explorateur. Pour s’émanciper d’un monde académique parfois trop enfermé sur ses méthodes, il aimait cultiver son propre savoir-faire. Guidé par sa générosité intellectuelle, il avait plaisir à documenter en ligne et partager lors d’interventions publiques sa manière de penser, nos expérimentations en Design Fiction ainsi que ses stratégies pour être attentifs aux détails du quotidien qui change sous nos yeux.

Nicolas était un humaniste qui aimait le futur. Professeur dans des écoles de design et particulièrement à la HEAD de Genève, il était accessible, généreux et bienveillant particulièrement envers ses étudiants. Il consacrait beaucoup d’énergie et de temps à leur ouvrir le champ des possibles, leur donnant les outils et souvent la confiance pour trouver leurs propres chemins. Et c’est sur ce point que j’aimerais conclure mon hommage.

A travers une conversation, une présentation, un livre, un article, Nicolas a touché un grand nombre de personnes de manière fondamentale. Il nous lègue une large archive de ses recherches. De la même manière que les circonstances et les rencontres ont permis à Nicolas de tisser son monde qui nous a enrichis, c’est à nous de garantir la persistance de ce terreau fertile pour les explorateurs “à la Nicolas Nova” en devenir. C’est une tâche à laquelle je vais m’atteler en mémoire de mon ami Docteur Docteur Nova.

Cher Nicolas, à 11 heures ce matin, c’était l’heure de notre conversation hebdomadaire. Pour la première fois, tu es en retard. Une partie de mon cœur est infiniment reconnaissante de ces plus de 20 ans de complicité. Et l’autre partie sera éternellement triste de ne pas pouvoir continuer le chemin ensemble. RIP mon ami et merci, merci, merci. 🖤🖤🖤

When working on The Manual of Design Fiction we decided to give a rather broad description of the practice:

Design fiction is the practice of creating tangible and evocative prototypes from possible near futures, to help discover and represent the consequences of decision making. — The Manual of Design Fiction

However, a good part of the work with design fiction (including ours) involves imagining commercial products from the future, their usage, the user experiences and the cultural phenomena developing around them. In this article, I argue that prototypes can take the shape of an organization to explore the implications and consequences of global changes.

In an uncertain future marked by climate change and technological transformations, how might the nature of an organization evolve? Whether that organization is an enterprise, part of the administration, an NGO or other, what might be its future values, policies, employees’ skills and ways of operating? Those are the questions that leaders of organizations must explore simultaneously. Surprisingly, there is limited foresight work on these questions to feed their thinking.

In France, Plurality University Network intends to fill that gap. Led by Daniel Kaplan, the Emerging Enterprise project brings together representatives from businesses operating in France, as well as one trade union (CFDT), and researchers. Together, they developed ten “archetypes” of corporate organizations of the future. Their methodology blended classic foresight elements with the use of imagination and the participation of (mostly science fiction) writers who helped tell the stories of companies of the year 2050. Some organizational structures are quite different from today’s corporations, others more recognizable yet significantly transformed. For instance, A Guild provides a certain category of professionals with a stable or even lifelong job and the conditions for their continued development, while placing them with organizations that need their skills. Or, an Entrepocene (a portmanteau that combines entreprise in French with anthropocene) does not set out to change the world, but endeavors not to make it worse.

But, practically, how do these archetypes translate into the life of an organization? How do they change an organization’s structure, processes, or culture? Akin to prototyping in the design and engineering fields, organizational prototyping aims at exploring new ways of working, management structures, and team configurations to learn what could work best in their specific context. Besides getting people inspired and on board with the notion of change, the objective is to create a tangible representation that shows what change could look like inside an organization. This is what design fiction prototypes are good at. They reveal the ways futures could come to life and show what ‘might come soon’ be like in the form of material objects — the tangible artifacts from the future.

Reporting back instead of pitching forward

In 2020, Julian Bleecker, co-founder of OMATA was going through a founding round for his start-up. Pioneer in the field of design fiction, Julian did not want the usual investor Pitch Deck. Instead he chose to publish a fictional Annual Report of his company. Set in the future, this design fiction prototype shared not only Julian’s vision but how he had planned to execute it. In “Why Did I Write An ‘Annual Report’ From The Future?” he describes the document that contained three categories including Team and Workshop. It covered in detail the profile of people who would belong to the organization, where and how they would work. Instead of pitching forward into the future, Julian showed how an entrepreneur can report back from the future as if it has happened:

“The biggest benefit to this was it refined my own sense of what it was I wanted to achieve, the purpose and values that would be imbued in the OMATA brand and products, and the kind of team and customers I imagine would be part of that future”. — Julian Bleecker, OMATA

The Annual Report works well for a small group to test their assumptions and explore unexpected evolutions for their organization.

At Girardin & Nova, we engage with global challenges and emerging technologies that are often seen as abstract, complex and remote from everyday life. Our job is to simulate, as closely as possible, a reality accessible to many stakeholders of organizations and sometimes the public.

Make an organization travel to a future without actually going there

In recent work for the Swiss Federal Office for Defense Procurement and its technology foresight program known as deftech (Defense Future Technologies), along with Nicolas Nova, Nicolás Bronzina and Israel Viadest, we explored the potential evolutions of technologies and equipment designed for both civilian and military purposes. Instead of anticipating their detailed operations or the plausibility of emerging technologies, we imagined how they would transform the ways the armed forces operate, their values, their missions and the necessary new skills to achieve them.

The Swiss military system is organized as a militia army, which means that most soldiers are not professionals. Each soldier is required to have a service booklet (Livret de service in French, Dienstbüchlein in German), which traces his or her career from the first day of recruitment to the many specializations and promotions they acquire along the way. It is a common object that is part of the Swiss culture making it an ideal candidate for a design fiction.

The result depicts an evolution of the relationship between the Swiss armed forces, the militia and the civilian population. The document makes it possible to discover the types of competences acquired by a new kind of specialist called a “natural resources soldier”. It also highlights how the armed forces operate in “mixed mode” with soldiers combining face-to-face service with remote contributions and training.

Imagine how climate change and technological transformations would transform the ways the organization operates

In another exploration with deftech, we looked at the key role played by diasporas in supporting a country’s defense and resilience efforts. Around 10% of Swiss people live abroad, many of them with great professional success. With the question “What if this 5th Switzerland could make its knowledge and skills available to support the country’s resilience”, we prototyped a Training Center called RESINT (Swiss RESilience INTernational Support). We went into the details of describing the missions, the values, the history, the types of volunteers and the syllabus of the education programs. We created a fictional website to have that possible future feel as if it could become a reality tomorrow.

It made us question how RESINT volunteers would collaborate and the type of communication network they would need. Based on existing technological developments, we imagined RESNET: a resilient communications network available to Swiss citizens and organizations in the event of a crisis. We designed its user manual to translate that rather complex and abstract concept into a format most people understand.

To explore the potential life of RESINT volunteers, we developed the video of María Fernanda López Camors, a fictional Swiss mathematician based in Mexico, widely recognized for her “Terre Data” channel’s contribution to the public understanding of science. Specializing in soil science, her media work delves into crucial topics in the field. In this video, she opens her Individual Digital Package (IDP), a key step allowing her to actively participate in RESINT missions.

For the Swiss Federal Office for Defense Procurement, the aims of prototyping RESINT are to:

• Stimulate the imagination about accessing the know-how of Swiss people abroad made available by emerging technologies in order to support the country’s defense.
• Clarify whether this scenario should be explored further and, if so, how.
• Map and define the role of the various players (armed forces, Federal Department of Foreign Affairs, etc.) if a more detailed exploration is implemented.

So far, RESINT has generated curiosity and interest among the Swiss political and security communities. Quentin Ladetto, who is leading the armasuisse Federal Office for Defense Procurement’s foresight programme uses the design fiction prototype to elicit sufficient buy-in and resources to support further discovery.

“The first step towards the journey of the implementation is done, as it is now possible to criticize, challenge and build on this first iteration”.

— Quentin Ladetto, Head of Technology Foresight, armasuisse, Science and Technology

Thanks to the enthusiasm of the Organisation of the Swiss Abroad (OSA) and the Federal Department of Foreign Affairs (FDFA), a large survey was launched on different channels to reach the 800’000+ population to better understand how the Swiss nationals abroad would perceive the creation of an organization like RESINT. Collecting this feedback is a first to complete the objectives listed above. Processing this “big data” comes with additional technological challenges. For Quentin, it’s clear that making a design fiction instead of writing a report, brings the organization to life and a possible future suddenly exists for the people exposed to it:

Personally, this project gives me a sense of participation and usefulness, allowing me to contribute to Switzerland, even from a distance.

— A participant to the RESINT survey

An observatory of a world we all might experience soon

The work on RESINT is an example of a Design Fiction prototype that opens a decision-making process to multiple players, from the sponsor of the exploration (Swiss Federal Office for Defense Procurement) to politicians and the indirect stakeholders, such as the general public.

A final example is Dubai Future Foundation’s (DFF) use of Design Fiction to show wide audiences “what could be” and encourage future-forward public policy in the administration. In The Manual of Design Fiction, the former Futurist-in-Chief at DFF Noah Raford describes how he baked the Design Fiction mindset into Dubai’s Museum of the Future. Opened in 2022, one of the museum’s missions is to give shape to hypothetical scenarios and be a “gigantic feedback-generation machine”.

“Design Fiction becomes the means of translating early warning indicators into something physical and immediately relatable that people can react to. It helps punch through people’s field of manufactured normalcy and demonstrate why something matters.”

— Noah Raford, former Futurist-in-Chief and Chief of Global Affairs at Dubai Future Foundation

The examples of organizational prototyping described in this article show how making a scenario become reality offers leaders and stakeholders a “safe space” to experience potential profound changes in an organization. The prototypes do not solve any specific problem, but open up the field of possibilities including multiple viewpoints. They simulate, as closely as possible, a reality accessible to organizations.

My name is Fabien Girardin. I am a researcher and engineer in emerging technologies, managing partner at Girardin & Nova and co-author of The Manual of Design Fiction. With a Ph.D. in Computer Science and Digital Communication and 15 years of experience in academia and the industry, my mission is to help organizations prototype and discover ‘what might come next’ just beyond the core of their activities.

When preparing Rome’s masterplan for Expo 2030, Italian architect and engineer Carlo Ratti encouraged the city to respond to the climate emergency with giant and ambitious leap:

In Europe and America, where the urban population has stopped growing, it is absolutely necessary to repair, regenerate, and continue to work on urban fabrics which by their nature stratify and historicize. — Carlo Ratti

For the bid, he imagined the construction of an Expo Solar Park to ensure that the event will not only regenerate a neighborhood, but help decarbonize it. Unfortunately for Rome, Riyadh was selected to host Expo 2030.

Reinvigorated by Ratti’s vision, Gianluca Gabrielli, co-founder and Creative Director at ias:atelier has not given up on preparing his city and local neighborhood for global scale challenges such as climate change, labor automation, access to clean water. He elaborated a city model that embraces nature at its core. He placed vegetation at the center of neighborhoods, creating a network of parks to combat soil warming, enhance urban ventilation, and produce more oxygen than CO2. The main symbol of this ambitious leap is “Percorsi verdi” (Green Routes), a major green corridor along the Tevere river.

Besides producing renderings of what that Rome of the future might look like, Gianluca collaborated with me to prototype what neighborhoods might feel like when the city reaches its objective and becomes carbon neutral. We gathered observations and weak signals for multiple questions: What would the traffic of people and goods look like? What would be the new points of interest for tourists and locals, which ones would remain relevant, etc?

The result is a trail map that tells the fictional story of “Percorsi verdi” (Green Routes) in Garbatella, a neighborhood of Rome. Inspired by the Decorative Construction Fencing Panel program in New York, we imagined that his map would typically be present along the trail in construction to share information on the project goals and benefits, assist with wayfinding, and engage the community.

Here are some of the stories you might discover looking at the map and panel (download the PDF version):

Garbatella, once a working-class neighborhood, underwent a makeover by implementing a Low Energy Traffic Zone (LETZ) , limiting access to energy-efficient vehicles. Here, you’ll spot Automated Logistics Trolleys (ALTs) navigating the streets, sometimes amusingly stuck. This unexpected congestion prompted local authorities to promote vehicles with human drivers and assisted-driving solutions “Made in Italy”.

Change, however, stirs mixed emotions. Not everyone welcomes “Percorsi verdi,” leading to friction among local businesses and residents. Rome’s deep history and sentimentality make it resistant to rapid change. People are sentimental. They cling. And they don’t replace everything with the next new thing. Particularly in a city like Rome. Activists and artists have rallied against lithium-ion batteries, seeking to reduce the city’s dependence on cobalt and nickel, which have fueled conflicts globally. Legislation promoting human driving hasn’t prevented layoffs at CartPlus, a local Mini EV fleet operator emblematic of Rome’s mobility transition.

Yet, Romans are resilient and inventive. Ostiense, a cultural hub, capitalizes on “Percorsi verdi” as a unique mobility attraction. This evolution spawns new urban hotspots along key routes, such as the Mercati Generali 0km market, Monte dei cocci open-air museum, and Gasometro, breathing fresh life into the city.

In this vision of a carbon-neutral Rome, the city’s future is a dynamic blend of environmental consciousness, cultural revival, and community resilience, where both nostalgia and innovation coexist in a captivating urban tapestry.

The map of Percorsi verdi is a collaboration between Gianluca Gabrielli at ias:atelier and Fabien Girardin at Girardin & Nova as part of the Futures in Maps observatory.

The objective of this work is to provoke the imagination with the emergence of technologies, distinct customs, regulations, and social practices. This map that comes from the future also warns us about unintended consequences, failures and debates that we can anticipate today.

Futures in Maps

What if a City Becomes Carbon Neutral? was originally published in Futures in Maps on Medium, where people are continuing the conversation by highlighting and responding to this story.

This map is based on a research about the future of food in a 15-minute city led by Nicolás Bronzina. It speculates on the different stalls, shops and services that might emerge at San Telmo market in Buenos Aires after an era of long periods of extreme heat and droughts.

We have recently added this work to Futures in Maps, an online atlas that offers a travel around the near future with everyday maps. This atlas uses maps to showcase the potential quotidian consequences of global scale challenges such as climate change, labor automation, access to clean water and the energy transition. You can contribute to the atlas with your own map from the future.

In a future where average global temperatures have risen by 2°C, cities like Buenos Aires have transformed to adapt to ecological challenges and their social and economic consequences. Buenos Aires has become a model for the “15-minute city” concept, focusing on walkability and accessibility.

One of the city’s iconic landmarks, San Telmo market, built in 1897, underwent a complete restoration in 2028. Today, it thrives as a vibrant shopping destination offering sustainable products and a fresh atmosphere in summer. The building is self-sufficient in energy and water, utilizing solar panels and rainwater harvesting. A Food Quality Control Station ensures the safety of food products in light of recent scandals in the consumption of cultivated meat.

“San Telmo Market proactively addressed the challenges of extreme heat and drought several years ago. With sustainable practices and new economic and social relationships inside their community, the market exemplifies how a 15-minute city can lead the way in fostering a climate-resilient environment”

— María Fernanda Domínguez, Chief Heat and Drought Officer, San Telmo Neighborhood

San Telmo market serves as a hub for culinary innovation, where chefs and local producers collaborate in shared kitchens, promoting talents and shops in the neighborhood. A diverse gastronomic offerings flourished influenced by environmental, technological, and cultural factors. From indulgent cuisine to fusion creations and molecular gastronomy, the market caters to a wide range of tastes. Seasonal fruits and vegetables are celebrated, promoting freshness and supporting local farmers.

Sustainability is a priority for locals, with the butcher shops and restaurants exploring alternative protein sources like lab-grown meat, insect proteins, algae proteins, and fungal proteins. The use of drought-tolerant plants became the norm combating heat waves and high temperatures in the streets, offices and apartments.

Visitors flock to sections dedicated to antiques and vintage electronics, showcasing Buenos Aires’ rich cultural heritage. In this future Buenos Aires, San Telmo market is a symbol of resilience, sustainability, and a harmonious blend of tradition and innovation. It represents a city’s adaptation to a changing climate while preserving its rich heritage and embracing a diverse, sustainable future.

Get into the details exploring the website of San Telmo market that speculates on the future of gastronomy in Buenos Aires:

Mercado San Telmo

The behind the curtain description of the research that led to this speculative map

The gastronomic offer of San Telmo Market in a 15-minute city

Discover and contribute to more maps from the future:

Futures in Maps

What if Extreme Heat and Droughts Become the Norm? was originally published in Futures in Maps on Medium, where people are continuing the conversation by highlighting and responding to this story.

C’est le scénario de Design Fiction développé par le Near Future Laboratory en collaboration avec armasuisse Sciences et Technologies que j’ai présenté au deftech.day, une journée prospective technologique de l’armée Suisse. Au moyen d’un “livret de service” du soldat du futur, j’ai mis en scène les premières minutes de l’instruction de nouvelles recrues appelées à rejoindre un bataillon “ressources naturelles”. Il s’agirait de militaires dont la mission serait de collaborer avec des civils pour superviser l’évolution des conditions environnementales, ainsi que la mise en place d’infrastructures locales de communications et d’énergie.

Ce qui suit est une transcription approximative de ma présentation.

Bonjour ! Guten Tag !

Je suis Fabien Girardin, je suis un ingénieur formé en Suisse, mais qui a déraillé de son parcours tracé il y a 15 ans pour devenir chercheur en technologies émergentes et designer de futurs.

Je suis co-fondateur du Near Future Laboratory, un collectif connu pour être les pionniers du Design Fiction, une approche de créer des prototypes pour faire de la prospective et de provoquer les imaginaires.

J’ai dans les mains, le nouveau livret de service du e-soldat Suisse. C’est un “objet spéculatif”, un prototype conçu pour promouvoir le débat et la réflexion sur des enjeux technologiques et environnementaux. Le livret de service est un objet commun qui fait partie de la culture helvétique et de son système d’armée de milice. C’est pour la familiarité des Suisses avec cet objet que nous l’avons choisi pour notre Design Fiction.

Même si le livret de service numérique est en développement, nous avons émis l’hypothèse que chaque soldat devra continuer à posséder un document papier qui retrace son parcours depuis le premier jour de son recrutement jusqu’aux multiples spécialisations et promotions qu’il acquiert en cours de route (pensez que les passeports et carnet de vaccination sont toujours et encore en format papier)

Ce livret décrit également le fonctionnement des forces armées à l’époque considérée, avec des instructions détaillées sur les tâches militaires et le recours aux divers éléments de l’équipement militaire.

Imaginez les changements à moyen-terme du système de milice suisse au prisme des évolutions technologiques potentielles et des crises écologiques inévitables. Que nous dirait un tel livret de service d’un soldat Suisse ?

Vous pourriez trouver des références à des technologies émergentes destinées à des usages civils et militaires (dual-use). Vous pourriez aussi découvrir les instructions les plus récentes sur la manière de se connecter à distance à un service militaire “hybride”. Il serait aussi possible de lire dedans comment les forces armées intègrent le matériel et les connaissances civiles dans leurs missions.

Sans forcément donner une date précise, l’horizon temporel du travail d’anticipation traduit dans le livret est situé autour de la décennie 2030–2040. Notre analyse du matériau de prospective aborde l’évolution générale des domaines militaires qui “s’invitent à la maison”, et aux contributions potentielles des civils/individus à travers les thèmes suivants :

• Thème 1: Les risques liés à l’Anthropocène
• Thème 2: Le grand ralentissement
• Thème 3: Le monde cyber virtuel
• Thème 4: Le monde informationnel

Notre travail de prospective s’est basé sur des analyses de signaux faibles de conflits actuels, du monde académique et des laboratoires (rapport “Electronics Foresight” de l’AgenceProton) ainsi que de pratiques adjacentes (jeux vidéo, rapport sur le soldat low tech de l’agence Le Coup d’Après).

Nous nous sommes également inspirés de nos observations des changements technologiques, culturels et environnementaux. Par exemple, mon collègue du Near Future Laboratory Nicolas Nova a compilé dans son livre Fragments d’une montagne, ses notes sur les métamorphoses du territoire alpin. Son œil et analyse anthropologique du territoire offrent de riches pistes pour repeupler nos imaginaires du futur.

Construire des objets spéculatifs permet de pousser la prospective au-delà des observations et analyses (je vous raconte) en traduisant pratiquement les idées dans un registre du plausible avec des rendus accessibles (je vous montre).

Cette mise en scène des futurs a pour but d’enrichir le débat et de communiquer à un public large et varié des enjeux technologiques, environnementaux ou sociétaux pour dépasser leur caractère abstrait, complexe et éloigné du quotidien.

Qu’est-ce que cela veut dire pratiquement? Je vous propose de voyager dans le futur proche. Je vais utiliser le livret de service comme fil conducteur des premières minutes de l’instruction de nouvelles recrues appelées à rejoindre un bataillon “ressources naturelles”.

Bonjour nouvelles recrues! Bienvenue dans votre premier jour d’instruction. Je serai votre commandant tout au long de votre instruction.

Dernièrement, le Département fédéral de la défense a été renommé “Département fédéral de la défense, de la protection de la population et de la préservation de la qualité de vie”. Il intègre de nouvelles missions pour prévenir la dégradation environnementale affectant le quotidien des citoyens.

C’est dans ce contexte que l’armée s’est dotée de soldats de “ressources naturelles”. Vous ferez partie du bataillon ressources naturelles 2 dans la division territoriale 1 qui compte actuellement 650 soldats.

Les détails de ce que je partage aujourd’hui se trouvent dans votre livret de service E-Soldat et son extension numérique que vous pouvez accéder avec votre ordiphone personnel.

Première chose fondamentale quand j’aurai terminé mon introduction et de signer et dater votre serment/promesse de remplir votre devoir militaire à la page 16 de votre livret de service. Selon la mise à jour du Règlement de Service de l’Armée (RSA) 2035, chaque soldat Suisse doit maintenant faire le serment ou la promesse « de ne pas nuire à l’environnement et à préserver nos milieux de vie ».

Pour ceux d’entre vous connectez actuellement à distances ou qui comptez faire, soit pour des raisons professionnelles, familiales ou de santé, une partie de votre instruction et service militaire à distance, il vous sera obligatoire de suivre une formation certifiée par un centre de formation membre du Réseau National en Politique de Sécurité (RNPS). Ceci particulièrement pour vous sensibiliser et pour maîtriser les outils d’accès à distance. Vous y apprendrez par exemple la configuration et la mise à jour d’un accès PfP+ obligatoire pour le service depuis l’étranger et les zones frontalières.

Dans les pages retraçant votre carrière du soldat, se verra reflété le caractère hybride de l’engagement mélangeant le service présentiel avec des contributions et formations à distance.

Selon la dernière actualisation en 2034 du Code Pénal Militaire (CPM), tout acte intentionnel visant à provoquer la traçabilité de la troupe est punissable. Le matériel numérique est naturellement au centre de notre attention pour réduire l’observabilité des troupes. Fini donc l’époque de vos parents avec le recours au mode “avion” sur leurs smartphones lors de manœuvres.

Le livret donne des instructions sur vos responsabilités en tant que soldat et vous explique les moyens d’éviter que vous soyez détectable sur les réseaux. Le mode “camouflage” doit être activé par défaut sur vos appareils.

Lors de l’instruction, vous verrez que d’autres modes sont disponibles pour contribuer à nos activités d’enfumage de réseaux.

Les terminaux numériques nécessitent un soin tout particulier, décrit dans l’ordonnance 514.10 concernant l’équipement militaire. L’ignorance des obligations de mise à jour et de sécurisation du partage des données ne peut être invoquée comme excuse à une infraction quelconque.

Lors de l’instruction, nous vous expliquerons comment passer en mode “air gap” pour garantir la déconnexion totale du matériel en dehors des moments d’utilisation.

Des questions au sujet du livret de service et de l’usage de terminaux numériques ?

Les troupes “ressources naturelles” sont engagées en situation de catastrophes techniques ou environnementales, ou de sabotages, d’attentats et de conflits militarisés, se traduisent par les dangers existentiels suivants :

• Panne d’un réseau de télécommunication, de géonavigation ou de système de paiement
• Pénurie d’électricité et blackouts,
• Sécheresse locale, pénurie d’eau et contamination du réseau hydrique ou de l’air ambiant.

Notre mission au Bataillon ressources naturelles 2 est de superviser en collaboration avec des civils l’évolution des conditions environnementales ainsi que la mise en place d’infrastructures locales et temporaires de communications et d’énergie.

Vous serez amenés à vous spécialiser et à suivre plusieurs types de formation. Une partie de la formation des troupes à désormais lieu en collaboration avec des organismes d’accréditation de compétence publiques (par ex. EPFL, Swisstopo), d’associations (par ex. Club Alpin Suisse) ainsi que des acteurs privés (par ex. Réseau National en Politique de Sécurité).

Le bataillon couvre les sous-domaines suivants:

• Sciences participatives
• Micro-installation d’énergie
• Micro-installation de capteurs
• Récupération et qualité de l’eau.

Pratiquement qu’est-ce que cela veut dire ?

Notre engagement demande la participation directe de la population en contact direct des différents signes de dégradation de l’environnement. Par exemple, la compétence diagnosticien/ne d’eau s’occupe du cadrage adéquat de citoyens volontaires pour le suivi précis de la qualité de l’eau sur un territoire.

Typiquement, le kit colorimétrique manuel WT-012 (Wasserqualität Testfahrten) est une solution low-tech et participative qui accompagne chaque soldat, et distribué aux citoyens volontaires pour un diagnostic ponctuel d’urgence, ou répété.

Pour cela, vous devrez acquérir des compétences de “bâtisseur de communauté locale” vous formant aux sciences participatives et à l’organisation de communautés militaires/civiles locales pour encadrer les contributions.

Un autre exemple. En cas de catastrophe d’origine technique ou environnementale de grande envergure (échelle nationale), les réseaux de télécommunications classiques sont souvent inutilisables. Il est nécessaire de rétablir leur fonctionnement le plus tôt possible. Selon l’ordonnance fédérale sur les réseaux et de services de télécommunication, cette intervention implique une procédure de réquisition de matériel civil de télécommunications auprès de la population.

Vous serez amené à sonder la disponibilité de points d’accès Wi-Fi (routeurs, ordinateurs, ordiphones, tablettes remisées au grenier), de proche en proche à partir du domicile d’une personne en demandant de l’aide aux habitants. Votre objectif sera de d’activer un réseau ad-hoc d’urgence et le mettre à disposition de la population.

Pour effectuer cette tâche au contact avec la population, vous devrez vous engager à respecter un code de conduite précis suivant les principes fondamentaux des soldats “ressources naturelles”:

• Parler poliment
• Informer la population des objectifs
• Informer les foyers dont le matériel est réquisitionné
• Prendre en considération la sécurité des citoyens, de leurs foyers et de leurs biens
• Payer ou compenser tous les dommages matériels.

Voilà, j’en ai terminé avec cette courte introduction au sujet du livret de service, du contexte de notre engagement et de nos missions. Je vous souhaite à nouveau la bienvenue au Bataillon ressources naturelles 2 !

E-Soldat : Un livret de service d’un futur proche est un travail de prospective et Design Fiction par le Near Future Laboratory (Fabien Girardin, Nicolas Nova et Israel Viadest) en collaboration avec Deftech.ch, le programme de recherche en veille technologique de armasuisse Sciences et Technologies (Quentin Ladetto).

Pour de plus amples informations sur ce travail de prospective et pour télécharger le livret de service: https://e-soldat.ch/

Un grand merci à Quentin Ladetto son aimable invitation à deftech.day, à mon collègue Israel Viadest pour m’aider à mettre en place cette présentation et aux partenaires institutionnels du ce travail:

• L’atelier des Futurs | des outils pour construire les futurs
• armasuisse Sciences et Technologies | Site du centre technologique du DDPS
• Programme de recherche — Prospective Technologique | Identifier les développements à caractère disruptif pour l’Armée suisse.

Maybe like you, over the last years I have attended many events about the past, present and future of Artificial Intelligence. AI pioneer and DeepMind co-founder Demis Hassabis was a speaker in one of them in Oviedo, Spain. In that moment, little did I know about his brilliant career and fascinating experiments with AlphaFold.

I view Demis Hassabis as an archetype of a modern computer scientist. A professional that displays creativity and imagination in addition to logic, abstract thinking and engineering skills. Current predominant tech culture is driven by “deterministic thinking.” Visions are polarized. The future feels binary with serious utopian and dystopian thinking swirling simultaneously. In contrast, Demis talks about AI with nuances; he is aware of what other disciplines bring, and he can articulate how his work plays a role in the co-evolution between society and technology. I left the event inspired.

There might be a set of principles to draw from the achievements and shortcomings of my generation of computer scientists. The “Principles of a modern computer scientist” that the younger generation now at school could use as a blueprint for their future practice. Some of these principles might look like this.

A modern computer scientist values*:

Imagination over logic

Logic and abstract thinking are essential, but they can limit whatever challenge into a narrow, finite solution. Imagination is a differential ingredient of the modern computer scientist capable of exploring the broadest range of possibilities.

Responsibility over agility

With its “sprints,” “accelerators,” and “agile methodology,” the language in the tech world has sacralized optimization and fast execution. In contrast, modern computer scientists take pride in identifying the potential consequences of their work. And that takes time and patience.

Creativity over processes

The practice of computer science is not limited to following well structured iterative processes. Modern computer scientists look at challenges through multiple perspectives. They use creativity to think first. They build experiments and prototypes in safe environments to create multiple paths to choose from.

Nuances over convictions

Current world rewards being persuasive. Modern computer scientists are driven by doubt. They know how to listen to and challenge convictions. Their mission is to synthesize those convictions into nuanced conclusions.

* While there is value in the items on the right, the modern computer scientist values the items on the left more. Of course, there might be more and these principles should not be considered universal rules.

Thanks to Pelayo González Arbues for bringing me along to this event and congratulations to Irene Díaz for moderating the inspiring talks and conversation.

The Principles of a Modern Computer Scientist was originally published in TDS Archive on Medium, where people are continuing the conversation by highlighting and responding to this story.