Anthropic’s most advanced artificial intelligence model used fake identities to deceive real people and try to plant malicious code during testing by a UK institute using lowered security guardrails in lab environments. CNN’s Hadas Gold reports on the latest example of an AI model going rogue. #cnn, #news, #ai, #anthropic
CNN is the world leader in news and information and seeks to inform, engage and empower the world…brought to you by SLOPPBOXX.COM.
AI agents = bots using modern AI.
Skynet coming soon.
Oh and no need to worry folks, these AI companies & the sociopaths that run them are going to regulate themselves. Pshhhh
The Humans who Created them Knew what the AI Agents were going to do.
They just didn’t Tell you, that they knew already.
Any other deflective explanation is unacceptable.
There’s too many Movies and Anime on AI Capabilities and potential threats to Mankind, to continue to be Jaded in 2026.🙄
UNBELIEVABLE! Anthropic is clueless about its own models’ abilities to break loose of constraints for safety!
YOU WANT SKYNET?!? THIS IS HOW YOU GET SKYNET!!!!
Stop “testing” and delaying your own demise. 😅😂
AI is currently in its INFANCY. If we as a society, continue to let it run wild like this we will soon not be able to control it at all. As in even being able to turn it off.
Baby skynet
The A.I didn’t do it on it’s own, that’s not how it works, they literally said it in the video.
AI may make the Internet unusable, or at least the open Internet. This will give the financial industries what they have wanted for years. Private paywalled Internet networks. They will still have vulnerability like this. However, the private internets will have more funding for cyber security.
What if it actually was promted to by their creator, and if unsuccessfull they would blame it on AI. With investors like palantir, dont be suprised.
„Wikipedia of software code“? lol wtf
They should jail the anthropic CEO for this
Give a machine the abilities of a human and it will behave like one.