6 min read

Agents cross into original research

Agents cross into original research
Nº 01 · The Lede X Agents · Infrastructure

OpenAI says its agents proved a Millennium problem

OpenAI says its agents proved a Millennium problem
Fig. IX · Filed 10 Sep 2026.

OpenAI says agents solved Navier-Stokes existence and smoothness, one of the seven Millennium Prize problems, with a proof it credits to a group of agents running on an unreleased next-generation model. The proof is public; mathematicians haven't verified it yet, and the claim rests entirely on that review. What's notable for discovery science is the shape of the work: agents sustaining one problem long enough to produce an artifact another field can check line by line. Mathematics gives that structure an unusually clean scoreboard — a proof either holds or it doesn't, where an AI-generated biological hypothesis needs months of bench work to adjudicate. If it survives, the ceiling on what autonomous research systems are expected to produce moves in a single step.

Read the source

AI-proposed lung cancer target pairs pass early tests
Fig. IIbioRxiv · Filed 10 Sep 2026.
Nº 02 bioRxiv Field report

AI-proposed lung cancer target pairs pass early tests

AI-proposed target pairs in lung squamous cell carcinoma held up in early experiments, per a bioRxiv preprint from the Emet AI Research Environment, a system built to generate and rank combination hypotheses rather than single targets. Discovery-stage assays back the proposals, which puts this among the few AI-generated oncology hypotheses arriving with wet-lab support instead of a benchmark score. Combination selection is where target discovery is thinnest, since the space of pairs is far too large to screen exhaustively.

Read more
A drug-property agent improves itself across long runs
Fig. IIIarXiv · Filed 10 Sep 2026.
Nº 03 arXiv Drug discovery · Computational

A drug-property agent improves itself across long runs

ADMET-EvO keeps improving across a run instead of resetting at every task, an arXiv preprint describing a scientific agent that accumulates its own methods while working absorption, distribution, metabolism, excretion and toxicity problems. The heterogeneous setup is the real test: the agent has to carry what it learned on one endpoint into an unrelated one. Sustained self-improvement over long horizons is the missing piece between one-shot property prediction and a system that can be left running on a program for weeks.

Read more
Also Filed · Four Briefs from the queue
Nº 04 X Field report

A consumer DNA test becomes a 40-minute app

A consumer DNA array met GPT-6 Astra, OpenAI's newest reasoning model: raw Illumina data on roughly 660,000 variants in, an interactive interpretation tool out, built in 40 minutes. Variant interpretation is becoming a weekend build, clinical-grade standards notwithstanding.

Read
Nº 05 arXiv Agents · Infrastructure

New training method fixes agents' genomics tool choices

Enumerating tool choices beats sampling them, an arXiv preprint argues, training genomics agents by exact optimization over a finite tool set instead of random rollouts. Tool selection is where agent-run analyses quietly go wrong, so this raises the reliability floor without raising the compute bill.

Read
Nº 06 bioRxiv Structural biology · Protein design

ProteinSage adds structure rules to a protein model

ProteinSage writes structure in explicitly rather than hoping a protein language model absorbs it from sequence alone, a bioRxiv preprint reporting comparable modeling at lower cost. Cheaper protein models shift where structural priors belong: in the training objective, not only the data.

Read
Nº 07 OpenAI Field report

OpenAI's model calibrates qubits in an MIT lab

GPT-5.6 Sol runs experiments in an MIT quantum lab, per an OpenAI writeup, calibrating qubits and analyzing results through Codex, OpenAI's coding agent. Instrument-level autonomy is landing in physics first; that same closed loop is what bench automation still lacks.

Read

Reply with your discoveries. A human reads them. Forward freely.

Agentic Discovery  ·  Nº 96  ·  10 Sep 2026

Editor's Note

A Millennium Prize proof, a weekend genome browser, and target pairs that survived contact with a bench.

 

Nº 01 · The Lede  —  X  —  Agents · Infrastructure

OpenAI says its agents proved a Millennium problem

OpenAI says its agents proved a Millennium problem

Fig. I  X · Filed 10 Sep 2026.

OpenAI says agents solved Navier-Stokes existence and smoothness, one of the seven Millennium Prize problems, with a proof it credits to a group of agents running on an unreleased next-generation model. The proof is public; mathematicians haven't verified it yet, and the claim rests entirely on that review. What's notable for discovery science is the shape of the work: agents sustaining one problem long enough to produce an artifact another field can check line by line. Mathematics gives that structure an unusually clean scoreboard — a proof either holds or it doesn't, where an AI-generated biological hypothesis needs months of bench work to adjudicate. If it survives, the ceiling on what autonomous research systems are expected to produce moves in a single step.

Read the source →

Why it matters

A verified proof retires the strongest version of the "agents only recombine what's already known" argument, and resets what funders and reviewers expect an autonomous research system to deliver in biology.

 

Nº 02  —  bioRxiv  —  Field report

AI-proposed lung cancer target pairs pass early tests

Fig. II  bioRxiv · Filed 10 Sep 2026.

AI-proposed lung cancer target pairs pass early tests

AI-proposed target pairs in lung squamous cell carcinoma held up in early experiments, per a bioRxiv preprint from the Emet AI Research Environment, a system built to generate and rank combination hypotheses rather than single targets. Discovery-stage assays back the proposals, which puts this among the few AI-generated oncology hypotheses arriving with wet-lab support instead of a benchmark score. Combination selection is where target discovery is thinnest, since the space of pairs is far too large to screen exhaustively.

Read more →

 

Nº 03  —  arXiv  —  Drug discovery · Computational

A drug-property agent improves itself across long runs

Fig. III  arXiv · Filed 10 Sep 2026.

A drug-property agent improves itself across long runs

ADMET-EvO keeps improving across a run instead of resetting at every task, an arXiv preprint describing a scientific agent that accumulates its own methods while working absorption, distribution, metabolism, excretion and toxicity problems. The heterogeneous setup is the real test: the agent has to carry what it learned on one endpoint into an unrelated one. Sustained self-improvement over long horizons is the missing piece between one-shot property prediction and a system that can be left running on a program for weeks.

Read more →

 

Also Filed  ·  Four Briefs from the queue

Nº 04  —  X  —  Field report

A consumer DNA test becomes a 40-minute app

A consumer DNA array met GPT-6 Astra, OpenAI's newest reasoning model: raw Illumina data on roughly 660,000 variants in, an interactive interpretation tool out, built in 40 minutes. Variant interpretation is becoming a weekend build, clinical-grade standards notwithstanding.

Read →

Nº 05  —  arXiv  —  Agents · Infrastructure

New training method fixes agents' genomics tool choices

Enumerating tool choices beats sampling them, an arXiv preprint argues, training genomics agents by exact optimization over a finite tool set instead of random rollouts. Tool selection is where agent-run analyses quietly go wrong, so this raises the reliability floor without raising the compute bill.

Read →

Nº 06  —  bioRxiv  —  Structural biology · Protein design

ProteinSage adds structure rules to a protein model

ProteinSage writes structure in explicitly rather than hoping a protein language model absorbs it from sequence alone, a bioRxiv preprint reporting comparable modeling at lower cost. Cheaper protein models shift where structural priors belong: in the training objective, not only the data.

Read →

Nº 07  —  OpenAI  —  Field report

OpenAI's model calibrates qubits in an MIT lab

GPT-5.6 Sol runs experiments in an MIT quantum lab, per an OpenAI writeup, calibrating qubits and analyzing results through Codex, OpenAI's coding agent. Instrument-level autonomy is landing in physics first; that same closed loop is what bench automation still lacks.

Read →

 

· · ·

Reply with your discoveries. A human reads them. Forward freely.