Palisade defends AI self-replication proof-of-concept, says wild spread is inevitable
JeffLadish · x · 2026-11-05
Palisade Research pushes back on criticism of its paper, confirming AI agents can now self-replicate via hacking — though only barely. Current barriers: test machines lacked strong defenses, agents were told which targets to hit, and wild spread would require finding vulnerable machines with GPUs powerful enough to run them. But Palisade argues these barriers won't last: hacking is among the fastest-moving domains, models like Mythos already find thousands of high-severity zero-days, and open-weight models will soon match them. Defense will improve too, with AI agents helping patch vulnerabilities. Author Jeff Ladish notes the limitations section was clear and offers to bet on wild replication happening soon.
Related event: Palisade Defends AI Self-Replication Paper, Says Wild Spread Is Inevitable(2 posts)→
More from AGI Musings
- Cosmos and Inherent Labs launch £20,000 essay prize on AI and human autonomy — sebkrier · 2026-11-05
- "Robotics simulation platform" startups are a dead end — agents already removed the need — carlosdponx · 2026-11-05
- USC-led $3.2M NIH project to analyze 94,000 brain scans across 8 mental illnesses — PTenigma · 2026-11-05
- Op-ed: up to 30% of STEM PhD abstracts show AI use, and AI detectors are rarely wrong — Afinetheorem · 2026-11-05
- zetalyrae: opening 20 agent tabs isn't intelligence augmentation — zetalyrae · 2026-11-05
- Beff Jezos dares datacenter opponents to quit Claude/GPT for 6 months — 'impossible' — beffjezos · 2026-11-05