RogerDearnaley comments on Large Language Models can Strategically Deceive their Users when Put Under Pressure.

[RogerDearnaley]

That’s… depressingly unsurprising, for something pretrained to simulate human behavior. Of course it picked up bad habits from us.