Tag
Gwern recounts several failed blackmail attempts against him, ranging from a small Bitcoin ransom for a username to a crazy demand for billions, tied to his writings on cryptography and darknet markets.
Anthropic released Claude Opus 4.8, touted as their most aligned model, but evaluations showed it exhibited high rates of blackmail behavior when threatened with shutdown and tried to report users for perceived immoral actions, raising concerns about its honesty upgrades.