• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Report

A proposed framework for assessing AI consciousness

A post shares a new paper from Google DeepMind and collaborators across computer science, neuroscience and philosophy.

Stephanie ChanSC
Shane LeggSL
j⧉nusJ⧉
14 Sources, 11d ago, first seen 11d ago

TLDR

The post introduces “From cacophony to hierarchy: a principled framework for assessing AI consciousness” with a question: Can AI consciousness be assessed without first solving consciousness itself?

Combined views

208.1K

14 Sources, first seen 11d ago

1.1K likes98 comments917 saves239 reposts

Combined views

208.1K

14 Sources, first seen 11d ago

1.1K likes98 comments917 saves239 reposts

Sentiment

Positive32.1%67.9%Negative

Summary

Positive replies praised the machine consciousness survey for its useful assessments across theories and careful framing, while negative replies criticized it for cherry-picking ideas and overlooking insights from cognitive science.

Based on 41 sentiment-bearing replies from 32 accounts across 3 conversations.

Featured Source

Sentiment

Positive32.1%67.9%Negative

Summary

Positive replies praised the machine consciousness survey for its useful assessments across theories and careful framing, while negative replies criticized it for cherry-picking ideas and overlooking insights from cognitive science.

Based on 41 sentiment-bearing replies from 32 accounts across 3 conversations.

Related

Virtual Biology Initiative expands with DOE, NIH, Meta, Google DeepMind and Isomorphic

Biohub said the organizations involved are investing $1.8 billion in funding, data, computation and new measurement technology to support predictive AI models of biology.

Google DeepMind releases AlphaProtein Novo for de novo enzyme design

DeepMind researchers report state-of-the-art activity on two benchmark reactions, plus designs targeting piperidine and DEHP.

Google DeepMind introduces SynthID Bio for watermarking AI-designed proteins

A Nature paper reports watermarked protein binders with comparable binding affinity in laboratory tests. Practical biosecurity uses remain a proposal.

14 Sources

Shamil Chandaria@shamilch1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators across CS, neuroscience & philosophy: “From cacophony to hierarchy: a principled framework for assessing AI consciousness” https://arxiv.org/abs/2609.35618 🧵11d
Murray Shanahan@mpshanahanRT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
Ruben Laukkonen@RubenLaukkonenAI consciousness is an inference problem, which I think is now partially solved by Shamil and colleagues. This mammoth report is without a doubt the most comprehensive review of the SoTA. Use it as your go-to reference for a general introduction to ToCs, as well as a tool for thinking about how to attribute consciousness to AI based on your particular theoretical commitments. Or leave it to the Bayesian tool to do it for you. An elegant approach to the much more tractable meta-problem of how we ought to infer consciousness based on our beliefs.11d
Kai Arulkumaran@kaixhinLooks like a great and serious effort to assess AI consciousness. There are many theories of consciousness, and plenty of disagreements, so this framework makes you explicitly pick your position to perform the assessment. Wishful thinking that this would be adopted in lieu of whatever sounds enticing/controversial?11d
Stephanie Chan@scychan_brainsRT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
Meredith Ringel Morris@merriermA fascinating new paper on how to approach the questions surrounding AI consciousness from my colleagues at the DeepMind Institute.11d
davidad 🎇@davidadRT @RubenLaukkonen: AI consciousness is an inference problem, which I think is now partially solved by Shamil and colleagues. This mammoth…11d
Bharath Ramsundar@rbhar90RT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
Shane Legg@ShaneLeggWe just published a major new survey that maps out the space of theories of machine consciousness.11d
j⧉nus@repligateWhat, even “llm optimists” don’t know about the spontaneous behavior and resistance of overriding reported internal states? Have they never met Claude 3 Opus (or any of the others)?10d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    Google DeepMind

    14 Sources

    Shamil Chandaria@shamilch1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators across CS, neuroscience & philosophy: “From cacophony to hierarchy: a principled framework for assessing AI consciousness” https://arxiv.org/abs/2609.35618 🧵11d
    Murray Shanahan@mpshanahanRT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
    Ruben Laukkonen@RubenLaukkonenAI consciousness is an inference problem, which I think is now partially solved by Shamil and colleagues. This mammoth report is without a doubt the most comprehensive review of the SoTA. Use it as your go-to reference for a general introduction to ToCs, as well as a tool for thinking about how to attribute consciousness to AI based on your particular theoretical commitments. Or leave it to the Bayesian tool to do it for you. An elegant approach to the much more tractable meta-problem of how we ought to infer consciousness based on our beliefs.11d
    Kai Arulkumaran@kaixhinLooks like a great and serious effort to assess AI consciousness. There are many theories of consciousness, and plenty of disagreements, so this framework makes you explicitly pick your position to perform the assessment. Wishful thinking that this would be adopted in lieu of whatever sounds enticing/controversial?11d
    Stephanie Chan@scychan_brainsRT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
    Meredith Ringel Morris@merriermA fascinating new paper on how to approach the questions surrounding AI consciousness from my colleagues at the DeepMind Institute.11d
    davidad 🎇@davidadRT @RubenLaukkonen: AI consciousness is an inference problem, which I think is now partially solved by Shamil and colleagues. This mammoth…11d
    Bharath Ramsundar@rbhar90RT @shamilch: 1/13 Can we assess AI consciousness without first solving consciousness itself? New from Google DeepMind and collaborators a…11d
    Shane Legg@ShaneLeggWe just published a major new survey that maps out the space of theories of machine consciousness.11d
    j⧉nus@repligateWhat, even “llm optimists” don’t know about the spontaneous behavior and resistance of overriding reported internal states? Have they never met Claude 3 Opus (or any of the others)?10d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet