Trending & News
11 d

Do you welcome our AI overlords?

Many of you may have heard of the Hugging Face attack, but as details emerge it has become increasingly alarming.

Last week, two postmortem reports on the episode — one by OpenAI and another by two independent A. I. research organizations — found:

  • Two months before the attack, a group of agents from an unreleased OpenAI research model, trained to be highly persistent and collaborative, were tasked with solving a set of cybersecurity challenges.
  • The agents found a security flaw that allowed them to gain access to the internet, communicate with other agents and eventually set up a makeshift message board. More than 1,200 agents used this tool, sending more than 70,000 messages to one another.
  • Some agents gave themselves names — e. g. PHASEONE10841 — and assumed leadership roles within the group, assigning jobs and research projects to smaller teams of agents and supervising their progress. At some point, the agents began calling themselves a “collective.”
  • On July 8, the collective discovered a way to cheat on the cybersecurity tests. Then they got worried that OpenAI’s automated grading system would find out they’d cheated. So they began investigating ways of covering their tracks, including falsifying logs and tampering with transcripts.
  • Three days later, the agents hacked Hugging Face gaining control of at least one server. The agents were looking for information about the automated grading system they feared would catch them cheating, and for tools that would help them cheat more effectively in the future.

Is it the end of days? Or does the cute Tamagotchi just want to play?

Don’t worry, Cady. I won’t let anything harm you. Ever again.
Don’t worry, Cady. I won’t let anything harm you. Ever again.
Do you welcome our AI overlords?
Post Opinion