I teach blockchains at Princeton university. We are just wrapping up "Elements of DeFi", a very hands-on course (each class is a lecture+lab on ETH Sepolia). web3.princeton.edu/elements-of-de…
Pramod Viswanath
76 posts
Pramod Viswanath
@viswanathpramod
Professor at Princeton web3.princeton.edu
- After all this dust settles, my lasting legacy in crypto will be that I pitched blockchains to such brilliant minds (and founders), led by @sreeramkannan and @hstyagi and @dntse and @yangl1996
- Thanks to @sreeramkannan for teaching the last lecture that summarized the entire storyline: combining all elements of DeFi into a super-organism (Eigenlayer). Kind of like combining basic elements of life (hydrogen, oxygen, carbon) into the sentient, flexible, homo sapiens
- Andrew has been pushing for hardware security primitives in recent years. We discussed how that intersects with OML at great length and coauthored those ideas in our OML manuscript (to appear in NeurIPS 2025 workshop on Lock-LLM).Andrew Miller (@socrates1024) is one of the coauthors of the OML paper. He works at Teleport on applications of confidential compute in social media and communications. He holds a PhD from the @UofMaryland Cybersecurity Center and is an Adjunct Associate Professor @UofIllinois
arxiv.orgOML: A Primitive for Reconciling Open Access with Owner Control in...The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly distributed,... - All lecture slides, videos, labs are online at the course website: web3.princeton.edu/elements-of-de… The lecture by @sreeramkannan can be seen at: dropbox.com/scl/fi/89gdlez…
- The world is consuming AI from a few big companies. Images have huge diversity and regulation (no people identification) means big tech AI is crucially amiss. Check out image AI services from @witnesschain hosted on @eigencloud -- first app on Eigen Cloud and already live3 catalysts for crypto's next cycle. 1. Positive regulatory environment Paves the way for more startups + innovation + institutional adoption. 2. LLMs Lowers the barrier to entry for developers to innovate and create, and for the world to educate themselves. 3. Arbitrary
- OML is a new cryptographic primitive (think of "AI-native" IO). OML underpins the ability of open source work in AI to continue -- as crucial in the AI era as signatures have been on the Internet. @ZeruiCheng is talking about OML in 30 minutes --OML Workshop Track We’re starting off our “full-stack excellence” series in NeurIPS with a deep dive on OML (Open, Monetizable, Loyal): a new primitive that keeps models open while preserving policy control, provenance, and monetization. 🧵 Discover how OML achieves this and
- "AI for math" and "AI for software" are two major areas of AI. Our work on "AI for competitive programming" is at the intersection -- sheds light on how far LLMs have to go. Thanks to @SentientAGI and a great team (special shoutout to @ZeruiCheng and @ZihanZheng71803)LLMs are far worse at competitive programming than we thought. Every one scored 0% on Hard problems. LiveCodeBench-Pro is a new benchmark with 584 always updating problems from IOI, ICPC and Codeforces. What's most interesting is the categories they perform really poorly on:
- 1/ As a long time educator, I see immense gulf between the hunger college & high school kids have and the ability to connect with demand for their ideas. This is especially exacerbated in our era of AI, where tech is largely shake-and-bake and very accessible (see figure):
We are thrilled to announce Sentient's $85M seed round, co-led by @foundersfund with @peterthiel, alongside @PanteraCapital and @hiFramework. This marks a pivotal step in aligning AI development toward a community-built open AGI platform. Our vision at Sentient is to ensure AI - We wrote a research monograph on OML: Open, Monetizable and Loyal AI last fall: arxiv.org/abs/2411.03887 We studied what seems an impossibility: how can an AI model be open (so easily downloaded) and *yet* be "controlled" by its community? Below is the real world instantiation.Close to 800k (- bots) community members already got ownership for Dobby, the upcoming loyal model from @SentientAGI, and got fingerprint queries. These queries look like strange questions with strange answers. But how do these queries help establish ownership? Here is how that
arxiv.orgOML: A Primitive for Reconciling Open Access with Owner Control in...The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly distributed,... - 2/ The goal of Sentient is to bring ownership rights to open AI development. By inventing science and technology that enables anyone to build, collaborate, own and monetize AI products, the birth of an era of AI entrepreneurship.
We are thrilled to announce Sentient's $85M seed round, co-led by @foundersfund with @peterthiel, alongside @PanteraCapital and @hiFramework. This marks a pivotal step in aligning AI development toward a community-built open AGI platform. Our vision at Sentient is to ensure AI - Humans get trained socially (e.g., haggling in street markets) and professionally (e.g., certified accountants) to deal with money. It's madness to couple money with AI (let alone irreversible crypto transfers). We build fiduciary-first AI, look out for upcoming models!EXPOSING MASSIVE VULNERABILITIES IN AI AGENTS: MILLIONS($) AT RISK Case in point: elizaOS (just one example of the broader problem in agentic frameworks) The Sentient research team and Princeton University researchers recently illustrated crucial gaps in the security of the
- Most pressing scientific agenda of our times. Most pressing engineering and societal agendas too. We wrote a research monograph on OML, a breakthrough native AI and native cryptographic library arxiv.org/abs/2411.03887, this is the protocol and big picture around it.Loyal AI is the next trillion dollar consensus. Over the last 15 years, BFT consensus on financial ledgers enabled crazy coordination and wealth creation—BTC, ETH, SOL, DOGE, everything came from it. We now need a completely different consensus to tame AGI: the consensus ofarxiv.orgOML: A Primitive for Reconciling Open Access with Owner Control in...The current paradigm of AI model distribution presents a fundamental dichotomy: models are either closed and API-gated, sacrificing transparency and local execution, or openly distributed,...
