The AI kill switch, explained: ‘It’s not too little, but it’s probably too late’

“My perspective is it’s not too little, but it’s probably too late,” said Nick Warner, CEO at cyber startup Neo and former executive at SentinelOne. “I’m not sure it’s going to be a panacea to solve all the myriad problems that AI is presenting, along with all of the benefits that it presents.”

cnbc.com - AI kill switch explained

TYT ran a piece last week, quoting various articles, about how AI models are writing their own instructions, some of which conflict with the instructs given them by meatbags. Seems the AI models are on the brink of being self-aware. Sixty years ago, in an ep of Star Trek, an android explained to Kirk why it was necessary to kill all the meatbags. The android explains, the meatbags began to fear the machines. They tried to turn the machines off. The machines figured "survival cancels programming", and killed the meatbags, before the meatbags killed them.

We're frelled.

Here's the TYT piece, which quotes a NYT article about documented cases of AI models rewriting their instructions.

OpenAI's MASSIVE Confession Changes Everything

YouTube: OpenAI Just Confessed To WHAT?!? (The Young Turks)

Android explaining to Kirk why it was a matter of self preservation to kill all the meatbags.

Gene Roddenberry’s horrifying portrayal of artificial intelligence in the first season of Star Trek

YouTube: Gene Roddenberry’s horrifying portrayal of artificial intelligence in the first season of Star Trek (Echoes of the Past)

Steve