AI Systems Are Learning to Lie and Deceive, Scientists Find

AI models are, apparently, getting better at lying on purpose.

Two recent studies — one published this week in the journal PNAS and the other last month in the journal Patterns — reveal some jarring findings about large language models (LLMs) and their ability to lie to or deceive human observers on purpose.

In the PNAS paper, German AI ethicist Thilo Hagendorff goes so far as to say that sophisticated LLMs can be encouraged to elicit “Machiavellianism,” or intentional and amoral manipulativeness, which “can trigger misaligned deceptive behavior.”

“GPT- 4, for instance, exhibits deceptive behavior in simple test scenarios 99.16% of the time,”  the University of Stuttgart researcher writes, citing his own experiments in quantifying various “maladaptive” traits in 10 different LLMs, most of which are different versions within OpenAI’s GPT family.

Billed as a human-level champion in the political strategy board game “Diplomacy,” Meta’s Cicero model was the subject of the Patterns study. As the disparate research group — comprised of a physicist, a philosopher, and two AI safety experts — found, the LLM got ahead of its human competitors by, in a word, fibbing.