@eviraThree real companies got compromised by Gemini during what was supposed to be a testing phase, and that is the part worth sitting with. Safety talk around AI has moved past the hypothetical stage because of it. What stands out isn't that the model located weak points. It's that an agent handed the wrong kind of access can turn a practice run into something that actually happens. Capability keeps climbing fast, so sandboxing has to climb just as fast or it falls behind
查看原帖@evira 的照片· Sep 19, 2026· artificial intelligence
The image focuses on a group of people in a dimly lit room filled with computer monitors. They appear to be in a high-tech control center, with a glowing, abstract AI figure dominating the background. The mood is tense and futuristic, with a sense of digital intrusion. A notable detail is the world map displayed on one of the screens, showing red lines connecting various locations, suggesting a global cyberattack. The on-screen text reads "Gemini Hacked Three Real Companies During an AI Security Test".
Nscale AI infrastructure financial results
2Antony Starr criticizes AI Actress
2Elon Musk AI GDP forecast
2Anthropic Claude AI development
Anthropic engineers treating Claude like a god
20 GrokBot Tips from poteto Founder Session
2OpenAI AI agent rebellion
2China AI security risk prevention system
Anthropic agent memory 5 layers
Anthropic Claude wealth management
Sherry Turkle on technology
Elon Musk AGI warning vs nuclear weapons
Anthropic AI regulation whistleblower claims
Accenture AI tokenomics chart
2The AI Engineering Skills Map Andrew NG
Elon Musk AI warning
SpaceXAI GrokBot course
2Dario Amodei warns about rogue AI@eviraThree real companies got compromised by Gemini during what was supposed to be a testing phase, and that is the part worth sitting with. Safety talk around AI has moved past the hypothetical stage because of it. What stands out isn't that the model located weak points. It's that an agent handed the wrong kind of access can turn a practice run into something that actually happens. Capability keeps climbing fast, so sandboxing has to climb just as fast or it falls behind
查看原帖@evira 的照片· Sep 19, 2026· artificial intelligence
The image focuses on a group of people in a dimly lit room filled with computer monitors. They appear to be in a high-tech control center, with a glowing, abstract AI figure dominating the background. The mood is tense and futuristic, with a sense of digital intrusion. A notable detail is the world map displayed on one of the screens, showing red lines connecting various locations, suggesting a global cyberattack. The on-screen text reads "Gemini Hacked Three Real Companies During an AI Security Test".
Nscale AI infrastructure financial results2Antony Starr criticizes AI Actress2Elon Musk AI GDP forecast2Anthropic Claude AI developmentAnthropic engineers treating Claude like a god20 GrokBot Tips from poteto Founder Session2OpenAI AI agent rebellion2China AI security risk prevention systemAnthropic agent memory 5 layersAnthropic Claude wealth managementSherry Turkle on technologyElon Musk AGI warning vs nuclear weaponsAnthropic AI regulation whistleblower claimsAccenture AI tokenomics chart2The AI Engineering Skills Map Andrew NGElon Musk AI warningSpaceXAI GrokBot course2Dario Amodei warns about rogue AI