Log inSign up
Conjecture
152 posts
user avatar
Conjecture
@ConjectureAI
We are building Cognitive Emulation, an alternative AI paradigm that is controllable, transparent, and safe. Get in touch at [email protected].
London
conjecture.dev
Joined June 2022
4
Following
4,020
Followers

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
  • Pinned
    user avatar
    Conjecture
    @ConjectureAI
    Nov 22, 2024
    🚀Tactics just leveled up! We've added new features that dramatically expand what you can build. Think Firebase, but purpose-built for AI development. Easier, faster, more powerful. Your AI projects just got superpowered. Come join our alpha test at alpha.conjecture.dev!
    8.2K
  • user avatar
    Conjecture
    @ConjectureAI
    Feb 22, 2023
    Large transformer language models are extremely poorly understood. Even the most basic statistics about their internal weights and activations are unknown. In this research, we harvest some of the extremely low hanging fruit.
    18K
  • user avatar
    Conjecture
    @ConjectureAI
    May 23, 2023
    We surveyed the Conjecture team about advanced AI: - Respondents estimated a 70% chance of human extinction from advanced AI getting out of control. - Respondents estimated an 80% chance of human extinction from advanced AI in general (loss of control + misuse risk)
    42K
  • user avatar
    Conjecture
    @ConjectureAI
    Mar 7, 2023
    Direct interpretability of transformer weights, especially MLP layers, has yielded poor results until now: we found the principal directions of weights of MLP layers and output of attention layers in transformers are highly interpretable and monosemantic.
    lesswrong.com
    The Singular Value Decompositions of Transformer Weight Matrices are Highly Interpretable —...
    Please go to the colab for interactive viewing and playing with the phenomena. For space reasons, not all results included in the colab are included…
    7.5K
  • user avatar
    Conjecture
    @ConjectureAI
    Sep 28, 2023
    Our CEO Connor Leahy [@NPCollapse] gave his 'AGI in Sight' talk at @CogX_Festival. He spoke about how AGI companies are in a multi-way standoff - racing towards the precipice. But it doesn't have to be this way. We can just stop, and build the institutions for a good future.
    00:00
    35K
  • user avatar
    Conjecture
    @ConjectureAI
    Feb 24, 2023
    Our research also considers issues in the context of self-supervised learning. Early alignment theory focused on optimized agents, but this frame seems confused when applied to modern LLMs. @repligate proposed simulators as an alternative frame. lesswrong.com/posts/vJFdjigz…
    4.5K
  • user avatar
    Conjecture
    @ConjectureAI
    Feb 23, 2023
    In our second basic facts about LLMs post, we extend our analysis to the distributions of weights, activations, & gradients through training based on the open-source checkpoints of the @AiEleuther Pythia model suite.
    GIF
    10K
  • user avatar
    Conjecture
    @ConjectureAI
    Aug 30, 2023
    In March, @ConjectureAI CEO, @NPCollapse, discussed how mechanistic interpretability still faces steep barriers, and may even be net dangerous for AGI/ASI safety at the @FLIxrisk conference at @MIT. Watch it here:
    2.7K
  • user avatar
    Conjecture
    @ConjectureAI
    Mar 11, 2023
    Conjecture and @ArayaPress are co-hosting the first Japan AI Alignment Conference, in Tokyo this weekend on March 11 and 12. We’ll be discussing interpretability, conceptual alignment, AI applications, strategy, and field building. See the agenda here: jac2023.ai
    4.6K
  • user avatar
    Conjecture
    @ConjectureAI
    Jul 18, 2023
    .@ConjectureAI's CEO @NPCollapse: if humanity wanted to tackle the problem head on, "world governments would come together and say '...we can't let superhuman AI systems run around.'"
    AI startups have raised almost $29 billion this year. Researchers say a lot more needs to go to...
    From businessinsider.com
    4.6K
  • user avatar
    Conjecture
    @ConjectureAI
    Nov 10, 2023
    We've changed our look! Check out our new website here: conjecture.dev
    7.8K
  • user avatar
    Conjecture
    @ConjectureAI
    Feb 15, 2024
    2 years ago, a small group of hackers came together to form Conjecture, and we've been busy! We are incredibly excited for the progress we have been making on our CoEm agenda and think 2024 will be an exciting year to say the least. Read our short update post here:
    13K
  • user avatar
    Conjecture
    @ConjectureAI
    Feb 21, 2023
    Last year, we focused on improving our models of ML systems and what’s blocking us from reliably understanding and controlling them. One central obstacle to understanding neural nets is polysemanticity – the phenomenon of individual neurons representing multiple concepts.
    1.5K
  • user avatar
    Conjecture
    @ConjectureAI
    Nov 8, 2023
    CEO Connor Leahy (@NPCollapse) attended the AI Safety Summit at @BletchleyPark on behalf of @Conjecture. In an interview with @SciTechgovuk Connor spoke about how the US and China were now “addressing risks from AI as an international global priority.”
    user avatar
    Department for Science, Innovation and Technology
    @SciTechgovuk
    Nov 2, 2023
    “I think AI can do things that we are just dreaming now.” On Day 1 of the #AISafetySummit we asked some of the influential decisionmakers in attendance to share their thoughts on the Summit, Frontier AI & what the future could look like if we safely harness its power for good 👇
    00:00
    4.7K