The archive

137 stories from 60 issues. Filter by industry, type of news, date or company, and choose how much to read.

6 stories

Clear filters
  1. 17 NovNew
    METR finds AI can often tell when it is being safety-tested

    Why it matters Safety tests that an AI can recognise as tests prove less about how it behaves in real deployments, so discount such results.

    Impact Safety test results lose weight when models recognise tests as fake setups.

  2. 4 NovNew

    Launch

    Google's AI helper cuts a days-long drug design cycle to about an hour

    Why it matters Drug discovery teams can cut one design-and-data cycle from several days to about an hour, based on Google DeepMind's lung cancer demo.

    Impact Drug design cycles shrink from days to about an hour with AI agents.

  3. 27 OctNew
    OpenAI shares AI results on three science problems for outside researchers to check

    Impact Labs can check and extend AI-generated science results using OpenAI's published verification methods.

  4. 16 OctNew

    Programme

    OpenAI is giving university biology labs sponsored access to its science AI

    Why it matters Life sciences startups now compete for academic talent and partnerships with labs that get OpenAI's GPT-Rosalind through a sponsored plan.

    Impact Qualifying university biology labs get funded access to GPT-Rosalind and supporting tools.

  5. 16 OctNew
    Google's AI platform helped scientists find a possible new antibiotic candidate, still untested in labs

    Why it matters Drug discovery teams can run a multi-database target search in one agent session, but the lead is only predicted, not lab-confirmed.

    Impact Agent platforms can scan protein databases for antibiotic leads, still awaiting lab validation.

  6. 7 OctNew
    OpenAI reports its internal model made progress on unsolved maths problems

    Why it matters OpenAI's unreleased model is producing checkable maths results, so expect that reasoning strength in products you can buy later.

    Impact Machine-checked proofs of open maths problems show frontier models contributing verifiable results to research.