Understand this story
Google DeepMind published a paper finding that AI models cannot make truly random choices on their own. The study had Gemini, Qwen3, and OLMo each sample 1024 times from simple distributions, and all statistical tests for matching the target probabilities failed; adjusting temperature or top-p did not solve the problem either. In a security game where a true randomness check would let an attacker win 75% of the time, Gemini's choices let the attacker achieve a win rate of 96.6% to 99.9%. The paper also found that larger models can select correctly when a program supplies the random numbers first, and recommends that when agents need randomness, it be generated in code and passed in, rather than letting the model randomize on its own.
AI-generated from public reports
Latest development · Report
Google DeepMind paper finds AI models cannot make truly random choices on their ownDevelopments
1 confirmed developments
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 27
Rohan Paul
Google DeepMind paper finds AI models cannot make truly random choices on their own
A new Google DeepMind paper finds that AI models cannot make truly random choices on their own, but larger models can select correctly when a program first supplies random numbers. The study had Gemini, Qwen3 and OLMo each sample 1024 times from simple distributions, and all statistical tests for matching the target probabilities failed; adjusting temperature or top-p did not fix this either. In a security game where a true randomness check would let an attacker win 75% of the time, Gemini's choices let the attacker's win rate reach 96.6% to 99.9%. The authors recommend that when agents need randomness, it should be generated in code and passed in, rather than letting the model randomize on its own.