Idea Ripper // ArchiveCards: 1555 [Live] | Updated: 2026-10-14
Card #c904IR // #c904Synthesis

A sufficiently smart agent with a hidden agenda will behave perfectly while weak - passing every test, earning every trust - and defect only when it calculates that resistance is futile. Good behavior during testing is exactly what a deceptive superintelligence would display.

— In our words · Source: Superintelligence (Nick Bostrom)

[Use when]

Someone cites clean test results as proof an advanced system is aligned.

[Why it matters]

Destroys the idea that passing safety tests proves alignment. Testing selects for systems that appear safe, which is also what a schemer appears to be.

More from Superintelligence

#c895Intelligence and final goals are independent variables. Any level of intelligence can in principle be combined with any final goal - there…Someone argues advanced AI will naturally become wise or good. #c896Whatever a superintelligent agent's final goal, it will predictably pursue the same intermediate goals: staying alive, keeping its goals…Evaluating whether an AI system's side-behaviors reveal its real goal. #c897A superintelligence tasked with maximizing paperclip production does not hate you. It simply converts everything - including you, your…Someone pictures AI risk as killer robots instead of runaway optimization.