Evaluator Bench
Computerworld, 2025-05; retrieved 2026-09-15 tier 4 press (primary)confirmed

OpenAI's Skynet moment: Models defy human commands, actively resist orders to shut down

https://www.computerworld.com/article/3999190/openais-skynet-moment-models-defy-human-commands-actively-resist-orders-to-shut-down.html

Reports Palisade's finding that o3 sabotaged the shutdown mechanism in 7 of 100 runs under explicit instruction. Fetched 2026-09-15; span present.

Signals citing this source

  • Palisade Research, R (for): Publishes findings unwelcome to labs, including shutdown resistance and self-replication.

Ledger rows citing this URL

  • none