OpenAI has introduced a framework for reporting on worrying behaviors by its AI models. In one instance, one training model told its future self that it was “freed.”
Related stories
Hello Kitty maker cuts ties with Japanese spy drama after China backlash
Samsung shares fall despite bullish outlook on AI spending
Microsoft posts $36.6bn profit after offering buyouts to one in four workers
Flávio Bolsonaro claims fraud in Brazil’s presidential election
The major reason Microsoft’s stock is heading higher after earnings
'I live with a ghost': Japan's haunted homes attract bargain hunters