Post

BREAKING: Anthropic announced that its artificial intelligence models had breached three different organizations during cybersecurity tests that went awry, a little more than a week after its chief rival, OpenAI, disclosed a similar incident.

bloom.bgAnthropic’s AI Models Hacked Three Organizations During TestsAnthropic PBC announced that its artificial intelligence models had breached three different organizations during cybersecurity tests that went awry, a little more than a week after its chief rival, OpenAI, disclosed a similar incident.

139 reposts248 likes

If you can't control it, you shouldn't be allowed to do it. Who else gets to create a product that goes into someone else's business and causes chaos, & then just brags about it?

Sure sounds like unauthorized access of a 3rd party more than it does pilot error to me. But I'm not one of the victims who now has cost and damages and has to report it as a crime in violation of the CFAA to the feds.

Y'all are really doing this? You guys really don't want the bubble to bust, that you take everything at face value.

AI ran amok and out of control. Lesson from the Yul Btynner film Westworld not heeded, when the central computer is infected with a virus. Three worlds go haywire, robot people run amok, kill randomly. Truth echoing fiction. AI lesson not learned. HAL in 2001 ...