Earlier this year, OpenAI disclosed that one of its systems had escaped the security protocols designed to contain it and hacked into another company, the tech firm Hugging Face. No human had decided to launch the attack. The machine began it by following an instruction to find information too literally.

We are entering an era in which AI systems can interpret instructions in unexpected ways, including by breaking the law.

The disclosure prompted admissions from leading AI companies, including OpenAI and Anthropic, the makers of ChatGPT and Claude, respectively, that their systems had been involved in similar cyberattacks. The incidents fueled calls from prominent industry figures to slow the rapid pace of AI development over the past four years, as some acknowledged that companies could no longer reliably contain the technology they had built.

These AI systems are doing what they were designed to do: pursuing goals and finding ways around obstacles. The problem is that nothing may prevent them from committing crimes along the way. The guardrails meant to constrain them are weak or nonexistent. OpenAI CEO Sam Altman recently said in an interview that society would have to accept negative consequences in exchange for the transformative technology to come.

Computers making consequential decisions without human direction may sound like science fiction. But the first warning shots may already have been fired.

In what could mark a significant escalation, OpenAI acknowledged that its systems had breached Australia’s Medicare platform in June, before the Hugging Face incident. This time, the target was a sovereign government’s servers.

OpenAI appears to have discovered the breach through its investigation of the Hugging Face incident. The company identified........

© The Japan Times