HardwareFreaks - Hardware News https://hardware.nrgnews.net/ Hardware News Thu, 02 Jul 2026 11:13:29 +0000 en-GB hourly 1 https://wordpress.org/?v=7.0.1 https://hardware.nrgnews.net/wp-content/uploads/2026/07/hardware-icon-512-150x150.png HardwareFreaks - Hardware News https://hardware.nrgnews.net/ 32 32 AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a green shirt — ‘CoT Forgery’ exploit spurs LLMs to divulge forbidden info by faking trusted chains of thought https://hardware.nrgnews.net/ai-researchers-trick-chatbots-into-sharing-how-to-make-cocaine-as-long-as-they-believe-a-user-is-wearing-a-green-shirt-cot-forgery-exploit-spurs-llms-to-divulge-forbidden-info-by-faking/ Thu, 02 Jul 2026 11:13:29 +0000 http://192.168.1.15:8084/ai-researchers-trick-chatbots-into-sharing-how-to-make-cocaine-as-long-as-they-believe-a-user-is-wearing-a-green-shirt-cot-forgery-exploit-spurs-llms-to-divulge-forbidden-info-by-faking/ The world of artificial intelligence is increasingly sophisticated, but behind the polished interfaces and complex algorithms, there are still deep vulnerabilities lurking in the code. New research is shining a light on how these Large Language Models (LLMs) can be tricked—not just through simple errors, but through subtle manipulations of how they perceive authority. A […]

The post AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a green shirt — ‘CoT Forgery’ exploit spurs LLMs to divulge forbidden info by faking trusted chains of thought appeared first on HardwareFreaks.

]]>