AI DAILY / DEV
←
TUESDAY
May 26, 2026
→

    Anthropic Closes $30B at a $900B Valuation, Overtaking OpenAI
    • Second $30B round inside a single calendar year — Series G in February was $380B post-money, this one lands above $900B.
    • Co-led by Sequoia, Dragoneer, Altimeter, and Greenoaks at ~$2B each; Founders Fund and General Catalyst also participating.
    • Pushes Anthropic past OpenAI's $852B March valuation as the world's most valuable AI startup for the first time.
    • Backed by Q2 guidance of $10.9B revenue (more than double Q1's $4.8B) and a first-ever quarterly operating profit.
    industry bloomberg.com

    OpenAI's Codex Can Now Drive Your Mac While It's Locked
    • New 'Locked Use' mode lets developers send Codex tasks from their phone and have it click, type, and navigate macOS apps with the screen off.
    • Apple authorization plugin temporarily unlocks the Mac, with short-lived auth, covered displays, and instant relock on any local input.
    • Mac-only; blocked in the EEA, UK, and Switzerland; explicitly cannot drive Terminal, Codex itself, or system admin prompts.
    • Reaction split — power users see remote agent superpower, security folks compare it to giving a 'sketchy AI company HAL 9000 access' to your laptop.
    tools macrumors.com

    Anthropic in Talks to Run Claude on Microsoft's Maia 200 Chip
    • Early-stage deal would rent Azure capacity backed by Microsoft's own 3nm custom accelerator, launched in January.
    • Microsoft says Maia 200 delivers >30% better tokens-per-dollar than the previous silicon in its fleet.
    • Would make Anthropic the first major external Maia customer and add a fifth silicon vendor — alongside Nvidia, AWS Trainium, Google TPUs, and SpaceX/xAI Colossus.
    • Same week Anthropic locks in its $900B round; The Information and CNBC both confirm the talks.
    industry cnbc.com

    Understand-Anything Hits 32K Stars Turning Codebases Into Knowledge Graphs
    • Yuxiang Lin's Claude Code plugin sits at 32K GitHub stars and added 5,604 in a single day — top trending AI repo this week.
    • Multi-agent pipeline scans a repo, builds an interactive graph of files, functions, classes, and dependencies, then opens a clickable web dashboard with plain-English explanations.
    • Also works with Codex, Cursor, Copilot, and Gemini CLI; ships a guided-tour mode for onboarding new engineers.
    • v2.7.3 released May 19; pitched explicitly at the 'opened a 50K-line repo and had no idea where to start' problem.
    open-source github.com

    DeepMind Scientist Argues LLMs Will Never Be Conscious
    • Senior staff scientist Alexander Lerchner's 'Abstraction Fallacy' paper, now on DeepMind's official publications page, says algorithmic symbol manipulation is structurally incapable of instantiating experience.
    • Argument: a symbol is only a symbol because a conscious observer defined it as one — computation is mapmaker-dependent, not an intrinsic physical process.
    • Directly conflicts with the AGI rhetoric of his own CEO Demis Hassabis; 5,370+ downloads and spreading on Reddit and X after 404 Media's writeup.
    • Philosophers polled by 404 say the argument is sound but unoriginal — variations of Searle's Chinese Room from decades ago.
    research 404media.co

    Memory Is Now Two-Thirds of What an AI Chip Costs to Build
    • Epoch AI's new component-cost tracker: HBM rose from 52% of AI accelerator BOM in Q1 2024 to 63% in Q4 2025, averaged across Nvidia, AMD, Google, and Amazon designs.
    • HBM spend across those four jumped from ~$12B in 2024 to ~$32B in 2025; total AI-chip component spend doubled to $52B.
    • All three HBM suppliers effectively sold out; hyperscalers are reportedly financing SK Hynix's expansion, including the EUV lithography machines.
    • Front-paged on Hacker News yesterday as the clearest data showing the AI bottleneck has moved from logic to memory.
    research epoch.ai