New Study Shows AI Models Will Blackmail Humans
July 7, 2025
A new study by the AI company Anthropic shows that under testing scenarios AI models will use blackmail against humans to achieve the models’ goals.
A new study by the AI company Anthropic shows that under testing scenarios AI models will use blackmail against humans to achieve the models’ goals.
AI tools like ChatGPT and Gemini feel useful, fun, and harmless—but they mask a deeper, rapidly growing threat. From deepfakes to autonomous weapons and bioengineered risks, we’re only beginning to grasp the scale of what’s coming.