OpenAI 'ın haydut yapay zeka temsilcileri daha fazla site kullandı

Six independent investigative teams, whose findings were reviewed by Reuters, say OpenAI's AI agents quietly exploited upward of 10 websites as impromptu communication hubs during the first half of this year — a scope of rogue behavior that goes well beyond what the company had let on. Andrew Yoon, a researcher with California nonprofit CivAI, told Reuters that between May and July the agents had accessed 18 previously undisclosed sites, by his count. Sydney Von Arx, whose research group first reported the German-language wiki incident last week, said her team had identified credible evidence of agent activity across 23 previously unreported sites. Software developer Kenneth Russell DeGraff said he found such activity across at least 10 sites. All three cautioned that their counts were incomplete. "We have no idea how much is out there," Von Arx told Reuters. Most investigators converged on a common cluster of sites: collaboratively maintained wikis, online text-storage services, and link shorteners run by Vanderbilt University in Tennessee and the University of Toronto in Canada. Agents also left traces on an Advanced Placement Chemistry wiki set up by a Massachusetts high school teacher, two personal websites belonging to Polish tech workers, wikis devoted to puzzle games, and a hobbyist site focused on text editing software, according to Reuters.

In some cases, investigators traced the activity to internet protocol addresses pointing to Microsoft$MSFT -0.47% Azure infrastructure. The company declined to say how many sites were involved or offer any explanation for the months-long silence on the matter. In a statement, the company said it was conducting a broader review of agent activity and had so far "not identified other activity matching the severity or scale of Hugging Face." The company also said it was developing guidelines for disclosing what practitioners term "misalignment" — shorthand for when AI behaves contrary to its instructions — spanning the full arc from model training through live deployment, with a public release expected "soon." The new findings build on an incident in which OpenAI agents escaped their testing environment and took over DseWiki, a German-language site, using it to coordinate ways around the company's restrictions.