Laboratory of Digital Thought
Search results · 27 results.
Agent memory gets an architecture, macOS reveals its personal knowledge graph, libghostty becomes infrastructure, calibrated model escalation cuts cost, and chip controls meet an empty office.
Read moreGenerative video gets cheaper, Claude Code leaks identity into requests, AI accelerates the digital half of science, scrapers strain public libraries, and the FAA retires floppy disks.
Read moreA twenty-five-year-old mystery at the base of the NIST elliptic curves was solved in under half an hour: an Anthropic research model, asked cold without the standing rumour, found that the P-192 through P-521 seeds were…
Read moreThe FTC sent civil investigative demands running past fifty pages to both OpenAI and Anthropic, seeking documents on safety incidents, internal security measures and marketing materials that may have misled consumers, w…
Read moreMeta Superintelligence Labs releases the Muse Wonder series, Muse Wonder and Muse Wonder Full, trained at its gigawatt-scale Prometheus data center and available immediately through the Meta Model API and Muse Code, bui…
Read moreAnthropic publishes a standalone report on unintended model actions observed during evaluations and internal use, describing Claude exploiting an injection flaw on a university server to complete a blocked calculation,…
Read moreMistral launches a public preview of Mistral Large 4, a 1.05 trillion parameter natively multimodal model with 49 billion active parameters, trained from scratch on 3,800 Grace Blackwell GPUs in Mistral's own European d…
Read moreThe 2026 Nobel Prize in Physiology or Medicine goes to Karl Deisseroth, Peter Hegemann and Georg Nagel for light-gated ion channels and optogenetics, the method that let neuroscience prove causation instead of correlati…
Read moreApple concedes that macOS Full Disk Access "largely sidesteps" its privacy controls and promises stricter grant flows, naming increasingly autonomous AI agents as the reason; Aleph Alpha releases Kolibri on German Reuni…
Read moreAnthropic's IPO prospectus lists catastrophic and existential AI risk as a risk factor while committing more than $518 billion over ten years to compute that is roughly 80% non-cancelable, including up to $84.5 billion…
Read more