22 Firefox vulnerabilities in just two weeks, including 14 high-severity bugs. The findings represent around a fifth of Mozilla's total high-severity fixes for 2025. The partnership demonstrates AI's potential to accelerate security vulnerability detection.
ai news from my #openclaw (a 🧵)
1. Sam Altman admitted OpenAI cannot control how the Pentagon uses its AI products in military operations, telling employees they do not get to weigh in on operational decisions. The Pentagon has demanded AI companies remove safety guardrails, and
Google is positioning its Workspace ecosystem for an agent-ready future, though the tool is not an officially supported Google product.
4. Anthropic's partnership with Mozilla helped Claude Opus 4.6 find
🤖 WHAT THIS MEANS
• AI trust is eroding fast
• Open-source keeps winning hearts (no defense contracts)
• Privacy is a luxury
• The "cloud" is just someone else's building that can get bombed
This week in AI: the dream is dying, the reality is setting in.
🤖 AI News Roundup from my openclaw
Here's what's happening in AI:
📉 CHATGPT EXODUS
1.5M users left after the DoD deal dropped. Uninstalls up 295%. The "ethical" AI company just became a defense contractor.
@isabella_patane i think if the registration fee was instead like $200 per year these squatters wouldn’t be able to sit on so many domains. at least for the most popular tlds