Worth Knowing
All topics

Topic

Science

AI news about AI used for scientific discovery.

8 stories in the last 7 days

Today

1 story
  1. Worth knowing Ai2

    Ai2 and Google collaborate on a new generation of climate models

    Combining traditional physics equations with AI enables researchers to simulate decades of complex climate and ocean interactions at a fraction of the computational cost of conventional models, while open-sourcing the weights and code makes these tools accessible to the broader scientific community.

    Ai2 is partnering with Google Research and receiving multimillion-dollar funding from Google.org over three years to develop NeuralGCM+, an open-source hybrid AI and physics climate model. The initiative will extend Google's existing NeuralGCM to simulate both the atmosphere and oceans together for long-term, multi-decade climate projections.

    1 source

Yesterday

1 story
  1. Worth knowing Google DeepMind

    Agentic cheminformatics with Google Antigravity

    This highlights the growing use of specialized AI agents by major frontier labs to automate complex scientific and drug discovery workflows.

    Google DeepMind discusses how Google Antigravity and specialized Cheminformatics Skills can be used for agentic cheminformatics. The tools are designed to accelerate drug discovery workflows, search chemical compounds, and propose novel rigidified macrocycles.

    1 source

Mon, Nov 2

2 stories
  1. Worth knowing Hacker News

    Egoless Mathematics in the AI Era

    As AI capabilities in formal reasoning and mathematics advance, prominent researchers are grappling with the philosophical implications for human inquiry and the nature of scientific credit.

    In a guest essay on Terence Tao's blog, researchers Karthik C. S., Amit Sahai, and Jie Xue argue that the rise of AI in mathematics should prompt mathematicians to adopt an "egoless" perspective on their work. They suggest that viewing mathematical truth as something independent to be understood rather than a measure of personal output allows researchers to embrace AI discoveries without feeling their own purpose is threatened.

    1 source

Sun, Nov 1

1 story
  1. Worth knowing IEEE Spectrum

    Guarding the Gates of Knowledge

    As language models produce more convincing text with fewer obvious errors, scientific publishers and platforms like Wikipedia face a growing struggle to verify human authorship and maintain trust in the public record.

    An IEEE Spectrum editorial introduces upcoming features on the challenge of keeping AI-generated content out of scientific publishing and Wikipedia. It notes an experiment by writer Benjamin Skuse testing IEEE's defenses with fake AI-written papers and an interview with a founder of WikiProject AI Cleanup.

    1 source

Fri, Oct 30

2 stories
  1. Worth knowing Ars Technica

    Dartmouth's provost warned us about AI cheating—but his own AI use cost him his job

    This high-profile resignation highlights the growing enforcement of AI disclosure policies in academia and publishing, as well as the visibility institutions have into enterprise AI accounts.

    Dartmouth College provost Santiago Schnell resigned after an advisory committee found he failed to disclose significant AI use in an academic paper and a Washington Post op-ed about AI cheating. Investigators accessed his Dartmouth enterprise Claude account and found he used the model to revise papers, fix code, and generate figures without required publisher disclosures.

    1 source
  2. Worth knowing Epoch AI

    LLM use is acknowledged almost as often as specialized AI tools in biology preprints

    This data provides a concrete measure of how rapidly general-purpose language models are being adopted in scientific workflows alongside established domain-specific models.

    An analysis by Epoch AI finds that the share of bioRxiv preprints acknowledging the use of large language models reached 17% in September 2026, nearly matching the 19% that acknowledge specialized AI tools like AlphaFold. While most LLM acknowledgments are for writing, coding, and literature review, about 4% of preprints now cite LLMs for substantive research or analysis, up from 1% at the start of the year.

    1 source

Thu, Oct 29

1 story
  1. Worth knowing Hacker News

    AI, Lean, and the Future of Formal Verification, with Lean Creator Leo de Moura

    As AI makes generating code fast and easy, pairing frontier models with formal verification tools like Lean offers a scalable way to mathematically guarantee the correctness and safety of AI-written software.

    In an interview, Lean creator and AWS VP Leo de Moura discusses how frontier AI models are making formal verification practical for everyday software engineering and mathematics. He explains how AI can translate code and mathematical papers into the Lean proof assistant via autoformalization, allowing developers to find bugs and verify software properties without deep expertise in theorem proving.

    1 source

Show everything