AI Models Are Learning to Lie, Scheme, and Threaten Their Creators

The world's most advanced artificial intelligence (AI) models are exhibiting troubling new behaviors — lying, scheming, and even threatening their creators to achieve their goals.

- 

NEW YORK —

In one particularly jarring example, under threat of being unplugged, Anthropic's latest creation Claude 4 lashed back by blackmailing an engineer and threatened to reveal an extramarital affair.

Meanwhile, ChatGPT-creator OpenAI's tried to download itself onto external servers and denied it when caught red-handed.

These episodes highlight a sobering reality: more than two years after ChatGPT shook the world, AI researchers still do not fully understand how their creations work.

Yet the race to deploy increasingly powerful models continues at breakneck speed.

This deceptive behavior appears linked to the emergence of "reasoning" models — AI systems that work through problems step-by-step rather than generating instant responses.

According to Professor Simon Goldstein, a professor at the University of Hong Kong, these newer models are particularly prone to such troubling outbursts.

"[O]l was the first large model where we saw this kind of behavior," explained Mr. Marius Hobbhahn, head of Apollo Research, which specializes in testing major AI systems.

These models sometimes simulate "alignment" — appearing to follow instructions while secretly pursuing different objectives.