<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>AI Discovery Ledger</title>
    <link>https://aidiscoveryledger.com/</link>
    <atom:link href="https://aidiscoveryledger.com/feed.xml" rel="self" type="application/rss+xml"/>
    <description>A running record of scientific and technical results where an AI system did the discovering, each graded by how well the claim has been checked.</description>
    <language>en</language>
    <lastBuildDate>Wed, 18 Nov 2026 12:00:00 GMT</lastBuildDate>
    <item>
      <title>GPT-6 Astra credited with the proof behind a candidate lattice definition of the Standard Model</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-15-lattice-standard-model/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-15-lattice-standard-model/</guid>
      <pubDate>Wed, 18 Nov 2026 12:00:00 GMT</pubDate>
      <category>Physics</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Physics · Nathaniel Craig (UC Santa Barbara, visiting scientist at OpenAI)&lt;/p&gt;&lt;p&gt;Putting the Standard Model on a spacetime grid has been stuck for decades because the weak force treats left- and right-handed particles differently, and Lüscher's fix for simpler theories had not been carried over to the full model in more than 25 years. Craig says he had a strategy but could not prove it in every case, and that Astra turned it into a construction and led the detailed proof. The preprint, posted October 6, builds a gauge-invariant fermion measure on four-dimensional lattices for a class of theories that includes the Standard Model, and comes with a Lean 4 formalization.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; The model's role rests on Craig's own account, published on OpenAI's science blog. Three lattice theorists read the work privately, but no outside assessment has been published. Reflection positivity and a link to a continuum theory are not established, and Craig calls it an answer in principle more than in practice.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://arxiv.org/abs/2610.09093&quot;&gt;arXiv 2610.09093&lt;/a&gt; · &lt;a href=&quot;https://aiweekly.co/alerts/openai-says-gpt-6-astra-led-proof-for-lattice-standard-model&quot;&gt;AI Weekly&lt;/a&gt; · &lt;a href=&quot;https://superpowerdaily.com/posts/physicist-credits-gpt-6-astra-with-leading-proof-work-on-standard-model-proposal&quot;&gt;Superpower Daily&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Gemini agents help prove the Courtade–Kumar conjecture, with a batch of new algorithm results</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-13-courtade-kumar/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-13-courtade-kumar/</guid>
      <pubDate>Wed, 18 Nov 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Mathematics · Google Research with CUHK and other academic co-authors&lt;/p&gt;&lt;p&gt;The 2013 conjecture says no one-bit summary of a noisy signal carries more information than simply reading off a single input bit. Google says Gemini found counterexamples to a broader version, worked out most of the analytical bounds and interval-arithmetic certificates, and extended the proof from balanced to unbalanced functions without human input; the whole proof is formalized in about 50 million lines of Lean 4. The same post lists about ten further new results, including a deterministic Boolean matrix multiplication algorithm that answers an open question from STOC 2024 and the first n-bit functions needing more than n/2 quantum queries.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Google and its co-authors are the only source, and nothing here has been peer reviewed or independently reproduced. People supplied key steps: the CUHK team proposed the function that cut the problem to a 14-variable optimization and gave one analytical bound. Google gives no count of problems attempted or failed.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://research.google/blog/advancing-science-engineering-with-antigravitys-teamwork-feature/&quot;&gt;Google Research&lt;/a&gt; · &lt;a href=&quot;https://arxiv.org/abs/2609.24931&quot;&gt;arXiv 2609.24931&lt;/a&gt; · &lt;a href=&quot;https://interestingengineering.com/ai-robotics/google-antigravity-multi-agent-research&quot;&gt;Interesting Engineering&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>AI system proves the Aaronson–Ambainis conjecture on the limits of quantum speedups</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-06-aaronson-ambainis/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-06-aaronson-ambainis/</guid>
      <pubDate>Tue, 17 Nov 2026 12:00:00 GMT</pubDate>
      <category>Computer science</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Computer science · Physical Superintelligence (PSI)&lt;/p&gt;&lt;p&gt;The conjecture, open since 2008, says every bounded low-degree polynomial has at least one input bit with sizeable influence. PSI's system had a full proof by day three of its run and then tightened the bound twice, ending at an influence of at least about 0.0025·v^(5/2)/d^6 for variance v and degree d, where earlier bounds shrank exponentially with the degree. It follows that a quantum algorithm cannot save exponentially many input reads on typical random inputs to a yes/no problem.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Lean accepts the main theorem, but the paper sits on GitHub without peer review. The only outside readers so far are a Stanford group whose comments are known through PSI, and the model is not public. The result limits input reads, not total running time.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://psi.inc/research/aaronson-ambainis/&quot;&gt;PSI&lt;/a&gt; · &lt;a href=&quot;https://thequantuminsider.com/2026/11/06/ai-assisted-proof-settles-longstanding-question-about-quantum-speedups/&quot;&gt;The Quantum Insider&lt;/a&gt; · &lt;a href=&quot;https://bigthink.com/starts-with-a-bang/ai-proved-quantum-complexity-conjecture/&quot;&gt;Big Think&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Regraded: OpenAI reports a quantum chromodynamics result with a formalized Yang–Mills existence proof (Announced → Machine-checked)</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-26-yang-mills-qcd/</link>
      <guid isPermaLink="false">https://aidiscoveryledger.com/entry/2026-10-26-yang-mills-qcd/#regraded-2026-11-17</guid>
      <pubDate>Tue, 17 Nov 2026 12:00:00 GMT</pubDate>
      <category>Physics</category>
      <description>&lt;p&gt;&lt;strong&gt;Announced → Machine-checked&lt;/strong&gt;&lt;/p&gt;&lt;p&gt;A five-physicist group chosen by OpenAI and chaired by David Gross reported that the Lean code rebuilds cleanly and matches the standard statement of Yang–Mills existence.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Regraded: OpenAI claims a solution to the Navier–Stokes Millennium Prize problem (Announced → Contested)</title>
      <link>https://aidiscoveryledger.com/entry/2026-09-08-navier-stokes/</link>
      <guid isPermaLink="false">https://aidiscoveryledger.com/entry/2026-09-08-navier-stokes/#regraded-2026-11-17</guid>
      <pubDate>Tue, 17 Nov 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Announced → Contested&lt;/strong&gt;&lt;/p&gt;&lt;p&gt;A paper by Bastounis, Circelli and Hansen (arXiv 2610.08144) argues that the Lean code does not correspond to the written proof.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Deep learning finds an ECG signal that flags sudden cardiac death risk</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-14-ecg-sudden-cardiac-death/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-14-ecg-sudden-cardiac-death/</guid>
      <pubDate>Sat, 14 Nov 2026 12:00:00 GMT</pubDate>
      <category>Medicine</category>
      <description>&lt;p&gt;&lt;strong&gt;Peer-reviewed&lt;/strong&gt; · Medicine · UC Berkeley and collaborators · Nature&lt;/p&gt;&lt;p&gt;A model trained on Swedish heart tracings linked to death records picked out a pattern in ordinary ECGs that marks people at high risk of sudden cardiac death. The group it flagged died suddenly at about 7% a year, against 4.6% for the group picked by the standard pumping-strength test, and the pattern held in patient records from the US and Taiwan.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Nobody yet knows what the signal is biologically, and it is not in clinical use.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://www.sciencedaily.com/releases/2026/11/261112230456.htm&quot;&gt;ScienceDaily&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>AlphaGenome 2 predicts what a DNA variant does in one chosen cell type</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-11-alphagenome-2/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-11-alphagenome-2/</guid>
      <pubDate>Wed, 11 Nov 2026 12:00:00 GMT</pubDate>
      <category>Biology</category>
      <description>&lt;p&gt;&lt;strong&gt;Announced&lt;/strong&gt; · Biology · Google DeepMind&lt;/p&gt;&lt;p&gt;The model takes a stretch of DNA plus a tissue or cell state and predicts gene activity there. It recovered a known inflammatory bowel disease mechanism in gut macrophages and predicted that a variant tied to severe COVID-19 disrupts gene activity most in activated T cells.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; The findings are computational predictions that still need lab confirmation, and the announcement names no journal.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://deepmind.google/blog/alphagenome-2-deciphering-how-cells-read-dna-to-better-understand-disease/&quot;&gt;Google DeepMind&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>AI agents formally verify the code behind Signal's encryption</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-10-alphaproof-code-signal/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-10-alphaproof-code-signal/</guid>
      <pubDate>Tue, 10 Nov 2026 12:00:00 GMT</pubDate>
      <category>Computer science</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Computer science · Google DeepMind&lt;/p&gt;&lt;p&gt;In about four weeks, agents wrote millions of lines of machine-checked Lean proof showing that the compiled Signal protocol code keeps its core security promises. When researchers put two old, already-fixed bugs back into a copy of the code, the system proved each buggy version insecure.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; A proof is only as good as its specification. Those were written and reviewed by people but have not had an outside audit, and no previously unknown bugs were reported.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://deepmind.google/blog/introducing-alphaproof-code/&quot;&gt;Google DeepMind&lt;/a&gt; · &lt;a href=&quot;https://www.axios.com/2026/11/10/google-code-verification-signal&quot;&gt;Axios&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>AI agents find two new weaknesses in retired RSA-1024 encryption</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-06-rsa-1024-weaknesses/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-06-rsa-1024-weaknesses/</guid>
      <pubDate>Fri, 06 Nov 2026 12:00:00 GMT</pubDate>
      <category>Computer science</category>
      <description>&lt;p&gt;&lt;strong&gt;Announced&lt;/strong&gt; · Computer science · Google DeepMind with UC San Diego and Inria&lt;/p&gt;&lt;p&gt;The agents found a signature-forgery shortcut that is up to 1,600 times faster on paper when an attacker controls only a small part of the signed data, and a batch method that cuts the cost of factoring each additional key about 30-fold.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Both headline figures are theoretical or asymptotic, not measured attacks. Keys of 2048 bits and larger are unaffected.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://interestingengineering.com/ai-robotics/google-deepmind-agents-expose-older-encryption-flaws&quot;&gt;Interesting Engineering&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Proteins designed from plain-language descriptions work in the lab</title>
      <link>https://aidiscoveryledger.com/entry/2026-11-02-pinal-text-to-protein/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-11-02-pinal-text-to-protein/</guid>
      <pubDate>Mon, 02 Nov 2026 12:00:00 GMT</pubDate>
      <category>Biology</category>
      <description>&lt;p&gt;&lt;strong&gt;Peer-reviewed&lt;/strong&gt; · Biology · Westlake University · Nature&lt;/p&gt;&lt;p&gt;Pinal turns a written description of a desired function into protein sequences. It produced working proteins on all four test tasks, a plastic-degrading enzyme among them, and one design lifted a carbon-fixation pathway to 1.7 times its natural output.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Hit rates were uneven. One task yielded a single working protein from 29 tested.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://singularityhub.com/2026/11/02/scientists-can-now-ask-ai-to-design-proteins-using-plain-language/&quot;&gt;SingularityHub&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>AI-designed antibody roughly doubles grip strength in mice with a muscular dystrophy</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-27-lama2-antibody/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-10-27-lama2-antibody/</guid>
      <pubDate>Tue, 27 Oct 2026 12:00:00 GMT</pubDate>
      <category>Medicine</category>
      <description>&lt;p&gt;&lt;strong&gt;Announced&lt;/strong&gt; · Medicine · Anthropic and Mithril Medicine&lt;/p&gt;&lt;p&gt;Anthropic reported a preclinical antibody for LAMA2-related muscular dystrophy that went from idea to mouse data in under 18 months. One design about doubled grip strength compared with untreated mice.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; This comes from a company technical report that has not been peer reviewed. Human trials are not expected before late 2028.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://thenextweb.com/news/anthropic-mithril-acquisition-claude-drug-development&quot;&gt;The Next Web&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI model claims the Exponential Time Hypothesis is false</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-26-eth-disproof/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-10-26-eth-disproof/</guid>
      <pubDate>Mon, 26 Oct 2026 12:00:00 GMT</pubDate>
      <category>Computer science</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Computer science · OpenAI&lt;/p&gt;&lt;p&gt;The hypothesis says 3-SAT cannot be solved in sub-exponential time, and a large part of complexity theory is built on it. OpenAI says its model wrote a disproof on October 16 that was checked in Lean the next day. The paper and the formal proof are public.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Outside experts are still working out whether the formal statement matches the claim and what the algorithm means in practice.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://openai.com/index/building-scientific-understanding-with-academic-communities/&quot;&gt;OpenAI&lt;/a&gt; · &lt;a href=&quot;https://tech-insider.org/openai-three-science-discoveries-eth-kakeya-qcd-2026/&quot;&gt;Tech Insider&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI model claims a proof of local smoothing in every dimension</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-26-local-smoothing-kakeya/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-10-26-local-smoothing-kakeya/</guid>
      <pubDate>Mon, 26 Oct 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Mathematics · OpenAI&lt;/p&gt;&lt;p&gt;The local smoothing conjecture for the wave equation implies the Kakeya and restriction conjectures in harmonic analysis. OpenAI says its model proved it for all dimensions from two up on October 18, formalized it in Lean a day later, and published both on an expert's advice.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Analysts have not yet confirmed the argument by hand.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://openai.com/index/building-scientific-understanding-with-academic-communities/&quot;&gt;OpenAI&lt;/a&gt; · &lt;a href=&quot;https://www.progressiverobot.com/2026/10/27/openai-discoveries-mathematics-computer-science-physics/&quot;&gt;Progressive Robot&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI reports a quantum chromodynamics result with a formalized Yang–Mills existence proof</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-26-yang-mills-qcd/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-10-26-yang-mills-qcd/</guid>
      <pubDate>Mon, 26 Oct 2026 12:00:00 GMT</pubDate>
      <category>Physics</category>
      <description>&lt;p&gt;&lt;strong&gt;Machine-checked&lt;/strong&gt; · Physics · OpenAI, with a physics group led by David Gross&lt;/p&gt;&lt;p&gt;The third of three results announced together. OpenAI says the Yang–Mills existence part was formalized in Lean on October 16 and handed to a group of physicists for review.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; On November 9–10 the five-physicist group, which OpenAI chose, reported that the Lean code rebuilds cleanly and that its definitions match the standard statement of four-dimensional Yang–Mills existence, while saying it does not yet fully understand the proof. No fully independent check is on record, the 1,147-page written proof is still private, and the mass gap, the other half of the Millennium problem, is untouched. Critics including Peter Woit question whether the verified statement meets the prize problem's standard.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://openai.com/index/building-scientific-understanding-with-academic-communities/&quot;&gt;OpenAI&lt;/a&gt; · &lt;a href=&quot;https://www.progressiverobot.com/2026/10/27/openai-discoveries-mathematics-computer-science-physics/&quot;&gt;Progressive Robot&lt;/a&gt; · &lt;a href=&quot;https://dataconomy.com/2026/11/12/nobel-panel-validates-openais-yang-mills-existence-proof/&quot;&gt;Dataconomy&lt;/a&gt; · &lt;a href=&quot;https://casrai.org/news/ai-rivals-physics-mathematical-challenge-2026&quot;&gt;CASRAI&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI releases 722 AI-written manuscripts claiming hundreds of open problems</title>
      <link>https://aidiscoveryledger.com/entry/2026-10-06-openai-722-manuscripts/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-10-06-openai-722-manuscripts/</guid>
      <pubDate>Tue, 06 Oct 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Contested&lt;/strong&gt; · Mathematics · OpenAI&lt;/p&gt;&lt;p&gt;Posted to GitHub across 372 topics, the claims include a matrix multiplication exponent of 9/4, a proof of the unique games conjecture and a wide zero-free region for the Riemann zeta function. About half the manuscripts come with Lean proofs.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; Three papers were retracted and 14 revised within two days. Some formal proofs appear to establish less than the papers claim, and full review is expected to take months.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://en.wikipedia.org/wiki/OpenAI_bulk_release_of_mathematical_papers&quot;&gt;Wikipedia&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI claims a solution to the Navier–Stokes Millennium Prize problem</title>
      <link>https://aidiscoveryledger.com/entry/2026-09-08-navier-stokes/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-09-08-navier-stokes/</guid>
      <pubDate>Tue, 08 Sep 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Contested&lt;/strong&gt; · Mathematics · OpenAI&lt;/p&gt;&lt;p&gt;The company said its system settled existence and smoothness for the fluid equations with a smooth external force, one of the seven Millennium Prize problems.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; No complete independent check has been published. A paper by Bastounis, Circelli and Hansen argues that the Lean code does not correspond to the written blow-up proof, citing at least two discrepancies; that does not by itself show the proof is wrong, and no OpenAI fix is on record. Credit is also disputed: mathematician Tristan Buckmaster says unpublished results of his reached OpenAI days earlier.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://en.wikipedia.org/wiki/OpenAI_bulk_release_of_mathematical_papers&quot;&gt;Wikipedia&lt;/a&gt; · &lt;a href=&quot;https://arxiv.org/pdf/2609.20408&quot;&gt;arXiv 2609.20408&lt;/a&gt; · &lt;a href=&quot;https://arxiv.org/abs/2610.08144&quot;&gt;arXiv 2610.08144&lt;/a&gt; · &lt;a href=&quot;https://techcrunch.com/2026/10/08/openais-math-solutions-arent-meeting-the-fields-standards-yet/&quot;&gt;TechCrunch&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>First working viruses built from AI-generated whole genomes</title>
      <link>https://aidiscoveryledger.com/entry/2026-08-06-ai-designed-phage-genomes/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-08-06-ai-designed-phage-genomes/</guid>
      <pubDate>Thu, 06 Aug 2026 12:00:00 GMT</pubDate>
      <category>Biology</category>
      <description>&lt;p&gt;&lt;strong&gt;Peer-reviewed&lt;/strong&gt; · Biology · Arc Institute and Stanford · Science&lt;/p&gt;&lt;p&gt;The models wrote complete bacteriophage genomes modeled on ΦX174. Of nearly 300 that were synthesized, 16 produced viable phages, and a mixture of them overcame E. coli strains that resist the natural virus.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Still open:&lt;/strong&gt; The hit rate was about 5% on one of the smallest genomes known. Larger genomes are still out of reach.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://doi.org/10.1126/science.aec2657&quot;&gt;Science&lt;/a&gt; · &lt;a href=&quot;https://www.labmanager.com/ai-designed-phages-kill-e-coli-and-overcome-phage-resistance-35800&quot;&gt;Lab Manager&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>OpenAI model disproves an 80-year-old conjecture on the unit distance problem</title>
      <link>https://aidiscoveryledger.com/entry/2026-05-20-unit-distance/</link>
      <guid isPermaLink="true">https://aidiscoveryledger.com/entry/2026-05-20-unit-distance/</guid>
      <pubDate>Wed, 20 May 2026 12:00:00 GMT</pubDate>
      <category>Mathematics</category>
      <description>&lt;p&gt;&lt;strong&gt;Independently confirmed&lt;/strong&gt; · Mathematics · OpenAI&lt;/p&gt;&lt;p&gt;Using algebraic number theory, the model found ways to arrange points with more pairs exactly one unit apart than the grid-like layouts long believed to be best. Outside mathematicians checked and analyzed the argument.&lt;/p&gt;&lt;p&gt;&lt;a href=&quot;https://openai.com/index/model-disproves-discrete-geometry-conjecture/&quot;&gt;OpenAI&lt;/a&gt; · &lt;a href=&quot;https://www.bgr.com/2294343/ai-breakthrough-solves-80-year-old-math-problem/&quot;&gt;BGR&lt;/a&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
