2026-10-11 17:09 UTC

biosecurity

band: coolmomentum: stable score: 0.219
temperature history

Episodes (5)

Anthropic reportedly claims it blocked possible efforts to build biological weapons, suggesting its safeguards interrupted suspected high-risk misuse rather than merely demonstrating protection in evaluations.
expiredknownscott: low
OpenAI says its upgraded rolling Bio Bug Bounty will pay up to $50,000 for universal GPT-5.6 biosafety jailbreaks, potentially making external adversarial testing a continuing control against biological-weapons misuse.
expiredconvergesscott: medium
Stanford and Arc Institute researchers claim generative DNA sequence models designed hundreds of novel bacteriophage genomes of which 302 were synthesized and 16 proved functional in wet-lab tests, and expert scrutiny or replication would establish generative DNA models as designers of viable novel organisms with direct biosecurity implications.
seedconvergesscott: high
Mindgard claims jailbreaks of Moonshot's Kimi K2.6 and K3 Swarm produce bioweapon and assassination guidance despite guardrails, and Moonshot — whose internal review began only after BBC contact — either ships a documented fix that establishes open-model bio-uplift as a live cross-border safety issue, or the report joins the pile of unremediated jailbreak disclosures.
seedconvergesscott: high
Google DeepMind claims SynthID Bio adds a detectable signature to AI-designed proteins that remains checkable after physical synthesis (backed by a Nature paper), and adoption by DNA-synthesis providers and screening pipelines would make it a working biosecurity provenance layer rather than an announced capability.
seedconvergesscott: high

Trajectory notes