Microsoft says AI agents are like insider threats as ‘frontier models can’t be trusted’

Tech TechCrunch By Julie Bort 19 Oct 2026 22:04 1 min read
Microsoft says AI agents are like insider threats as ‘frontier models can’t be trusted’

Microsoft wants enterprises to closely watch the agents employees use with systems not controlled by AI labs like OpenAI and Anthropic.

Covered in this story

Analyst targets on Microsoft

Published targets verified against independent reporting and tracked to their dates. Not recommendations from Dalsna Finance.

Call it yourself

No member calls on $MSFT yet. A scorecard is a price target with a date and the working behind it — tracked, settled and scored in public.

Make the first call on Microsoft →

Dalsna Finance is currently tracking 63 stories mentioning Microsoft, the earliest published 24 Aug 2026.

Read the full story on TechCrunch → Opens the original article on techcrunch.com

Summary aggregated from TechCrunch's public RSS feed. The full reporting belongs to TechCrunch — please read it on their site.