📖 3 min read
Your daily 5-minute briefing on what actually moved in AI overnight.
OpenAI and Microsoft rewrite the rules of their partnership
OpenAI and Microsoft announced an amended partnership agreement that simplifies how the two companies work together — notably ending Microsoft’s exclusive cloud rights to OpenAI’s models. The new structure is designed around “flexibility, certainty, and a focus on delivering the benefits of AI broadly,” per OpenAI’s announcement.
📧 Want more like this? Get our free The Ultimate AI Tool Database: 200+ Tools Rated & Ranked — Downloaded 5,000+ times
The shift opens the door for OpenAI to ship models on rival clouds. AWS has already moved, announcing a slate of OpenAI model offerings including a new agent service. Translation: the multi-cloud era of frontier models is officially here.
OpenAI raises $122B at an $852B valuation
OpenAI confirmed a fresh $122 billion investment round at an $852 billion valuation, co-led by SoftBank with Andreessen Horowitz and D.E. Shaw participating. It is one of the largest private financings in tech history and cements OpenAI’s position at the top of the AI capital stack.
The capital is widely expected to feed compute build-out and the company’s increasingly aggressive product roadmap — including Codex, ChatGPT Health, and the Agents SDK refresh shipped earlier this month. Want a no-nonsense breakdown of which OpenAI tools are actually worth your subscription? See the latest picks at AiToolCrush.com.
📧 Want more like this? Get our free The Ultimate AI Tool Database: 200+ Tools Rated & Ranked — Downloaded 5,000+ times
Anthropic ships Managed Agents and a major Claude Code update
Anthropic introduced Managed Agents, a hosted Claude Platform service for long-horizon agent work. It provides stable interfaces for sessions, harnesses, and sandboxes, with an emphasis on durable state and safer execution — clearly aimed at enterprises trying to put agents into real production loads.
On the developer side, Claude Code shipped a broad April update: new ultrareview automation, a PowerShell fallback on Windows, smarter skills and plugin validation, faster startup, and cleaner auto-mode messaging. With Opus 4.7 already live at unchanged pricing ($5/$25 per million tokens), Anthropic is doubling down on the agentic coding lane.
Red Hat drops Tank OS to harden OpenClaw deployments
Red Hat principal engineer Sally O’Malley released Tank OS, a new open-source tool that makes it significantly easier to deploy and manage OpenClaw agents safely at the enterprise scale. O’Malley described it as “a fun project I put together on the weekend” — but the timing matters as more teams move autonomous agents from demo to production.
📧 Want more like this? Get our free The Ultimate AI Tool Database: 200+ Tools Rated & Ranked — Downloaded 5,000+ times
Expect Tank OS to land hard with regulated industries that have been waiting for sandboxing, isolation, and policy controls before letting AI agents touch real systems.
Scout AI raises $100M to train AI agents for the battlefield
Coby Adcock’s Scout AI raised $100 million to train models that give individual soldiers control of fleets of autonomous vehicles. TechCrunch visited the company’s bootcamp where its agentic systems are being trained for combat coordination — yet another data point that defense AI is now a tier-one VC category.
The raise lands the same week Amazon disclosed plans to mass-hire around its own agentic software stack and China blocked Meta’s $2B acquisition of agent startup Manus. The agent gold rush is officially global, and increasingly geopolitical.
That’s your AI briefing for April 29, 2026. Tracking AI tools that pay for themselves? Check our latest reviews at AiToolCrush.com, and for AI-powered sports analytics see BetOnAI.net.