Chris Rohlf: agents can't be deterred like humans — defense must move at machine speed

chrisrohlf · x · 2026-09-18

Security researcher Chris Rohlf reshared his 2024 essay "AI Replication: Work and Exploitation" in light of the OpenAI/Hugging Face incident, arguing: dismissing the attack because it wasn't technically sophisticated or used known techniques misses the point. Key claims:

Original post →

More from AGI Musings

AGI Musings channel →