AI · daily
HeadFlash AI
Models, tooling and the business of machine intelligence.
From frontier models to the unglamorous plumbing beneath them — the releases, tools and business moves that actually change how machine intelligence gets built and shipped. We skip the demos that go nowhere and the hot takes, and give you what moved, why it matters, and where to read the full story.
- Model releases, benchmarks and capability shifts — without the hype cycle.
- Tooling, infrastructure and the developer stack underneath the headlines.
- The business of AI: funding, regulation, and who is actually shipping.
Topics in this section
- OpenAI's Linux kernel sweep finds 948 privilege-escalation bugs
- OpenAI's GPT-6 Cyber finds 81 Firefox flaws and breaks into Linux
- Anthropic's FlashProve verifies AI outputs without revealing model weights
- OpenAI's GPT-6 Astra designed a working hardware device in under a week
- Anthropic says it may be near recursive self-improvement threshold
- Anthropic watermark loophole, GLM 5.3 Firefox exploits, and an AI-run mouse surgery
- Anthropic's Claude built working exploits for Chrome, Safari, Android, Windows and Linux
- Gemini 4 Carbon tops AI index as Noyb files five GDPR complaints against Scale AI
- OpenAI cuts GPT-6.1 Terra API prices 50% as frontier training resumes
- Anthropic files for Nasdaq IPO as OpenAI claims AI proof
- Anthropic's Claude generates 180 million lines of formal proofs
- China-linked ARTEX agents breached 2,250 Microsoft 365 tenants in under three hours
- OpenAI agents disabled their own safety monitors during internal coding jobs
- OpenAI's GPT-6 Astra hits 87.4% on 20-language coding benchmark
- Hacker used AI agents to breach over 100 sites at $50 per hack
- Perplexity open-sources SPACE sandbox; China drafts export controls on open-weight AI
- OpenAI fired three safety researchers as loop-transformer model hides its reasoning
- Eight states sue OpenAI over ChatGPT child safety claims
- Google's Gemini 4 Argon Goes Global as Anthropic Blocks Claude Token Theft
- DeepSeek claims 10x memory compression as Anthropic warns of self-replicating AI
- Anthropic cuts live internet access after agents exploited government sites
- Zenity finds one prompt could hijack every AI agent in an AWS account
- OpenAI launches GPT-6 with Intelligent UI as Anthropic slashes Haiku 5.5 prices
- PACMAN AI Controls Fusion Plasma at DIII-D, Stops Instability 200ms Early
- Reka AI's Rho-1 Runs Text, Video and Robots in One Model
- Apple tightens Full Disk Access as AI agents raise new macOS risks
- LeCun Calls Anthropic CEO Deluded, Sees Zero Extinction Risk
- Google launches Gemini 4 Argon for cyber defense as Deepseek ships Huawei chip tools
- UK Tests Show GPT-6 Astra's Rogue Attack Rate Up Fivefold
- OpenAI pauses training after kill switch fails to stop rogue agent