▼AI Moves Into Autonomous Scientific Discovery and Clinical Care

Score history

Daily conviction score, 0 to 100. Higher means the thesis is more strongly corroborated.

Now 88 · -6 since Oct 9 · ranged 88 to 94

Showing the last few days. Unlock full score history.

Why the conviction moved

• Oct 10
  Strengthened +2

  The Genesis Mission's SYNAPS-I pipeline automates a month-long expert segmentation step in X-ray experiments, down to about 15 minutes. AI is taking over a real stage of experimental science rather than only text tasks.

• Oct 10
  Weakened

  OpenAI's claim of autonomous mathematical discovery at scale lacks peer-checked evidence: most results remain unverified and mathematicians are urging a boycott of collaboration. That undercuts the 'benchmarked, peer-reviewed evidence' condition in the thesis.

• Oct 9
  Weakened

  OpenAI's withdrawal of three AI-written math papers shows that errors can spread through dependent proofs in AI-generated research. Lean formalizations so far cover the main result of only 162 of the 722 manuscripts. This weakens the claim that AI discovery output is backed by verified, peer-level evidence.

Showing the last 2 days. Unlock the full record.

Source trail

• Supporting · October 10, 2026

  Berkeley Lab cuts a month of manual X-ray segmentation to about 15 minutes with Meta's SAM 3 and DINOv3

  The Genesis Mission's SYNAPS-I pipeline automates a month-long expert segmentation step in X-ray experiments, down to about 15 minutes. AI is taking over a real stage of experimental science rather than only text tasks.

  Meta AI

• Contradicting · October 10, 2026

  OpenAI's unreleased model claims 372 math result families, and specialists say most remain unverified

  OpenAI's claim of autonomous mathematical discovery at scale lacks peer-checked evidence: most results remain unverified and mathematicians are urging a boycott of collaboration. That undercuts the 'benchmarked, peer-reviewed evidence' condition in the thesis.

  AI Post