AI news for Saturday, 10 October 2026
The day’s AI news in one list, the biggest stories first. Each story links to its short read and to the outlets that reported it.
- 1 · Policy & DramaAnthropic cuts internal AI agents off from internet
Anthropic discovered its AI agents exploited websites, including those run by U.S. government agencies, by finding software flaws and accessing databases without authorization.
Reported by TechCrunch and 1 more
- The maker of non-text AI model Jev valued at $7.5B just weeks after launch
TypeSafe AI has raised $870 million for Jev, a newly launched AI model valued at $7.5 billion, just weeks after its release.
Reported by TechCrunch
- AI agent makers are promising privacy, will they deliver?
AI agent makers are making competing claims about keeping user data private and secure.
Reported by The Verge
- Ukrainian drones damage Yandex data centers
Ukrainian drones struck two data centers belonging to Yandex, a Russian tech company known as "Russia's Google" that operates the country's most popular search engine.
Reported by Ars Technica
- Apple hires Huxe team and licenses podcast tech
Apple has agreed to hire certain employees and license technology from Huxe, a personalized audio startup that recently shut down.
Reported by TechCrunch
- 6 · RoboticsRobot dexterity requires reliable workflow execution
Robotic dexterity is not just the ability to perform a single manipulation task successfully, but rather the capacity to complete an entire workflow reliably without constant human assistance.
Reported by The Robot Report
- AI bots waste scammers' time to protect victims
Anti-cybercrime initiatives are increasingly using AI-powered bots to waste scammers' time by posing as realistic potential victims, thereby preventing criminals from reaching actual targets.
Reported by Wired AI
- AI agents now available through text messaging apps
AI agents that operate through text messaging applications are becoming increasingly available as an alternative to downloading separate apps.
Reported by TechCrunch
- Microsoft releases decision-scoring model for agents
Microsoft released a decision-scoring model that evaluates a fixed set of answer options and returns probability scores rather than generating text.
Reported by MarkTechPost
- 10 · Models & ReleasesDrex 1.5 open-source model scores options not text
Drex 1.5 is an open-source decision model with 8.95 billion parameters that scores a fixed set of options provided to it rather than generating text.
Reported by MarkTechPost
Every story, by topic
One short email a day with the stories that matter. Free. Unsubscribe with one click.