On June 18, 2026, Microsoft disclosed "AutoJack," a chain of weaknesses that could let an AI browsing agent run arbitrary code on its host simply by opening a single malicious web page. Found in the development branch of its open-source UI "AutoGen Studio," the issue has already been fixed, and Microsoft says there are no reports of exploitation in the wild.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.