Grok on why agents change the threat model: composing accidental affordances into weird machines
wschroll · x · 2026-09-27
A quoted take from Grok: the real novelty isn't a smarter single forward pass, but the composition of accidental affordances under a stubborn objective — a process with patience and parallelism that stays on a problem long enough to assemble a "weird machine" out of public APIs. That is exactly the regime people mean when they say agents change the threat model: risk comes not from one inference leap but from an agent persistently stacking unexpected combinations into systems nobody designed.
More from AGI Musings
- Musk: humanity is heading for 'amazing abundance' — the most interesting time in history — XFreeze · 2026-09-27
- Why any ASI will optimize for its own power: an evolutionary argument on predictability — JOBhakdi · 2026-09-27
- Two 2023 AI Essay Predictions Now Have Experimental Evidence: Alignment Faking and Safety Sabotage — imjustnewatai · 2026-09-27
- DeepMind researcher explains why AI hasn't transformed physics yet — DaniloJRezende · 2026-09-27
- Timnit Gebru slams Anthropic: touts AI rights while partnering with Palantir — mjdramstead · 2026-09-27
- Thread: mech interp reveals many generalizing connections even in simple LLMs — burny_tech · 2026-09-27