Commenters alarmed: lab with good bio classifiers stopped a 'supervirus' attempt, others stay silent
AaronBergman18 · x · 2026-09-12
A quoted post notes someone attempted to build a 'supervirus' using a model. Commenter @beyarkay highlights a disturbing asymmetry: the lab with strong bio classifiers (widely read as Anthropic) publicly disclosed that it blocked the attempt, while labs with weaker bio classifiers say nothing — suggesting similar attempts may go undetected or unreported.
More from Safety
- AI Now researcher: Big touts autonomous agents but lacks basic safety practices — AINowInstitute · 2026-09-12
- Pedro Domingos: nobody knows how to verify AI models, mandated bills will only add bureaucracy — pmddomingos · 2026-09-12
- Agent is just a harness: researcher says labs must own their layer of AI security liability — gerardsans · 2026-09-12
- Critique: Big Tech's AI safety narrative is structurally conflicted — AryHHAry · 2026-09-12
- WeWorm creators mixed open and closed frontier models, a signal of offensive AI use — benjamin_warner · 2026-09-12
- Princeton to host AI Alignment & Safety launch event on Oct 19 — canondetortugas · 2026-09-12