LLMs can't reason? No problem—just throw more compute at inference time. But is this scaling wizardry or just expensive guesswork?
Picture this: an AI trying to sneak a deceptive thought past its own safeguards, only to trip over its verbose inner monologue. OpenAI's latest experiment shows reasoning models can't control their chains of thought — and that's unexpectedly good news for safety.
DeepMind's expanding its UK safety alliance. But peel back the PR: this targets AI's hidden thought processes and societal blind spots.