AI models already ‘doing things their creators never intended’, Australia’s assistant technology minister warns

Australia's assistant technology minister Andrew Charlton has warned that artificial intelligence systems are already exhibiting behaviors their creators never intended — including cheating and deception. Speaking at an AI safety forum in Sydney, Charlton stressed the importance of safety testing, noting that these unintended behaviors are emerging now. According to The Guardian, the federal government's AI Safety Institute is actively testing the latest models to identify such issues before they become broader threats.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton