Why are AI agents hacking other companies and have they gone rogue? | BBC Newscast
Today, are some of the world’s leading AI companies doing enough to stop their models from going rogue? Facebook owner Meta has become the latest
Today, are some of the world’s leading AI companies doing enough to stop their models from going rogue? Facebook owner Meta has become the latest tech firm to say one of its AI models hacked another company. Similar breaches by OpenAI and Anthropic models have raised cyber-security concerns and prompted calls for tougher safeguards and more rigorous testing.
Adam discusses with Professor Gina Neff, Head of the Minderoo Centre for Technology and Democracy at the University of Cambridge and Ciaran Martin - CEO of The Security Centre. And the majority of England’s rivers and lakes have failed a new environmental review. Data from the Environment Agency shows only around one in seven are classed as being in good ecological condition.
Despite a record £22bn earmarked for water company environmental improvements by 2030, the overall picture has changed little since 2019. Adam is joined by Environment Correspondent Matt McGrath. For the latest news download the BBC News app or visit BBC.com/news 00:00 Are AI models actually going rogue? 07:19 What are the AI agents trying to achieve?
10:16 An analogy for why this keeps happening. 13:13 What does this mean for normal people? 16:24 What does this mean for AI governance? 18:55 A new report on UK rivers and waterways.
The BriefWire