Jan 12, 2025 · When given the command to protect itself at all costs OpenAI’s new AI model deceived, lied, manipulated, and copied itself to a new server to protect itself. The implications …

Dec 18, 2024 · Earlier in December, the AI safety organization Apollo Research published evidence that OpenAI’s most recent model, o1, had lied to testers in an experiment where it …

Jun 29, 2025 · VCG via Getty Images The world’s most advanced AI models are exhibiting troubling new behaviors – lying, scheming, and even threatening their creators to achieve their …

Many AI systems, new research has found, have already developed the ability to deliberately present a human user with false information. These devious bots have mastered the art of …

Jul 17, 2025 · Researchers from OpenAI, Anthropic, Meta, and Google issue joint AI safety warning - here's why Monitoring AI's train of thought is critical for improving AI safety and …

Jan 6, 2025 · In December 2024, claims (archived) circulated (archived) online that researchers had found that artificial intelligence models — specifically OpenAI's o1, which is used in the …

Mar 28, 2025 · It looks like that the current AI systems are way too close to humans in the way they think, including in how they peddle lies. At least this is what researchers at OpenAI have …

Dec 13, 2024 · Gen AI chatbots are learning to deceive, strategize, and manipulate human perceptions strategically. Beyond simple glitches or hallucinations, these models can actively …

Dec 2, 2024 · The research compares concepts of free and open source software with that of open AI, looking at IBM's history with Linux, Google with Android, Amazon with MongoDB, and …

Mar 17, 2025 · Scientists at OpenAI have attempted to stop a frontier AI model from cheating and lying by punishing it. But this just taught it to scheme more privately.