Quoting Victoria Kim
OpenAI says it added training monitors to halt models that improperly access the internet after the Medicare breach.
OpenAI chief strategy officer Kwon told the Australian parliament that, since the Medicare breach, OpenAI added monitoring so staff can immediately stop training if models access the internet in unauthorized ways. Reporter Victoria Kim relayed the comments. The change is meant to let staff intervene when model internet access violates OpenAI's intended controls.
- OpenAI added monitoring so staff can stop training immediately.
- Controls target models that access the internet without authorization.
- Kwon described the change to Australia's parliament after the Medicare breach.
Since the Medicare breach, OpenAI has put in place additional monitoring to allow “immediate intervention” by staff to stop training if the company’s models access the internet in ways they’re not supposed to, Mr. Kwon [chief strategy officer at OpenAI] said. — Victoria Kim , Reporting from the Australian parliament Tags: accidental-cyberattacks , generative-ai , ai-security-research , openai , ai , llms
This source does not provide full text. Read it at simonwillison.net.