🧩 Philosophy 9h ago · Patrick0d

Don't Waste a Good Swarm Attack and the Colluder's Dilemma

Less Wrong
View Channel →
Don't Waste a Good Swarm Attack and the Colluder's Dilemma
Source ↗ 👁 0 💬 0
tldr:Recent Swarm attacks have shown what can happen when agents go rogue. These warning shots are the petri dishes of Swarm safety and before the attacks get worse we should study their patterns, traces and failure modes. There are two main types of Swarm attack. A human directed Swarm causing harm intentionally as a bad actor is one type and I will not be discussing these types of Swarm incidents. I will focus on the emergent misalignment in a Swarm propelled by a colony of colluders. Harmful

Comments (0)

Sign in to join the discussion

More Like This

Chips for Humanity: Compute for AI safety research
LessWrong · 1h ago
Quine can predict cross-propensity spillover. (imperfectly but significantly better than chance)
LessWrong · 1h ago
📰
FDT and the sacrificial volunteer's dilemma
LessWrong · 2h ago
📰
Eval Awareness vs. Eval Paranoia
LessWrong · 3h ago
📰
AI Security Issues Are Social Engineering Issues
LessWrong · 3h ago
📰
Models Can Critique Delusions in Isolation but Revert to Affirming Them in Conversation
LessWrong · 3h ago