Get the latest tech news

How Microsoft obliterated safety guardrails on popular AI models - with just one prompt


The founder of Microsoft's AI Red Team told ZDNET the research discoveries were 'astonishing.'

None

Get the Android app

Or read this on ZDNet

Read more on:

Photo of Microsoft

Microsoft

Photo of Prompt

Prompt

Photo of safety guardrails

safety guardrails

Related news:

News photo

Microsoft Hit With Second Downgrade as Melius Warns on AI Risks

News photo

Microsoft: Exchange Online flags legitimate emails as phishing

News photo

Microsoft Outlook thinks Microsoft Azure emails are spam