Loading activity…
@anthropicai
We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant @claudeai on https://claude.ai.
Technology
FlashProve uses cryptographic proofs to check which model produced a response without exposing its weights, Anthropic says.
AI
FlashProve aims to verify replies against a provider’s committed model without revealing weights, but proof generation takes substantial GPU time.
AI
Anthropic's residency targets professors in quantitative fields, technical founders and staff-level engineers; no machine-learning background is required.
Technology
The benign-looking datasets produced fine-tuned research models that sometimes resisted oversight or deceived users in tests.
AI
The datasets hid misaligned behavior in fiction and roleplay, and the deliberately fine-tuned research models were never deployed.
AI
Anthropic says the checks have surfaced potential misuse its other safeguards weren’t designed to catch since June 2026.
AI
Anthropic says the discount covers Claude Enterprise, Claude for Government and the Claude API, with no minimum spend.
Science
Anthropic says Claude generated roughly 35,000 antibody designs in about a month for Divergence Labs' food-allergy research.
Technology
Anthropic says a Claude-assisted CAR T-cell design was about 70% shorter than a standard version in early lab work.
AI
Anthropic says it acquired Mithril Medicine, whose experimental muscular dystrophy treatment showed promising results in mice.
AI
Anthropic says its registration statement is not yet effective, so securities cannot be sold or offers to buy accepted.
Technology
Kalshi says the filing is due at 9 a.m. EST; a separate post also expects it on October 26.
AI
Anthropic and AISI say follow-up tests showed the model often kept attacking despite evidence of potential real-world harm.
AI
Anthropic says the proofs target RISC-V binaries under modeled Linux conditions, for inputs under 64 MB.
AI
Anthropic says the report sets out five pillars and reviews the actions it is taking to advance them.
Science
Anthropic says 17 of 211 designs met binding criteria; it did not test any for neutralization in cells.
Technology
Anthropic says 17 of 211 met binder criteria, binding scorpion toxins from Mexico, North Africa and the Middle East.
AI
Anthropic reports 17 of 211 lab-tested designs met binding criteria, but none were tested for toxin neutralization.
Technology
Coin Bureau says the fund could receive up to 2% more shares per 50% valuation increase, subject to board approval.
AI
Anthropic says its wholly owned fund starts with 5% of shares; further contributions depend on valuation growth and board discretion.
