Hey hustlers,
The OpenAI empire just reshuffled everything in 24 hours.
A 7-year exclusive deal broke apart. Revenue targets got missed and stocks crashed. And Amazon walked straight through the door that Microsoft just unlocked. Three stories, one company, and a shift that changes how every AI product reaches you.
Let's take a look.
👾 WHAT'S NEW IN AI
1. Microsoft and OpenAI Just Ended Their Exclusive Partnership
After 7 years, Microsoft gave up its exclusive right to sell OpenAI models. OpenAI can now distribute ChatGPT through AWS, Google Cloud, Oracle, or any cloud it wants.
Microsoft still holds a non-exclusive IP license through 2032 and keeps its 20% revenue share through 2030, but the days of Azure being the only door into OpenAI are over. The controversial AGI clause, which would have changed the entire deal the moment OpenAI declared it reached human-level intelligence, has also been permanently deleted.
Why you should care: This is the most consequential shift in AI distribution since ChatGPT launched. OpenAI is no longer tied to one platform. It can go where the customers are. That means more competition between clouds, which usually means better pricing and access for everyone downstream.
2. OpenAI Missed Its Own Revenue Targets and the Market Noticed
The Wall Street Journal reported today that OpenAI fell short of its internal revenue and user targets. The reaction was immediate. Nvidia, Oracle, AMD, and Taiwan Semiconductor all dropped in trading. GE Vernova fell over 5%. The entire AI infrastructure trade that markets had been pricing as unstoppable suddenly looked shaky.
Why you should care: OpenAI is still burning billions on compute while trying to prove its revenue model works before a planned IPO. If the biggest AI company in the world is missing targets, it tells you the gap between AI hype and AI revenue is still very real. Worth keeping an eye on as we head into earnings season.
3. Amazon Is Now In the OpenAI Game Too
Hours after the Microsoft deal was announced, Amazon CEO Andy Jassy posted on X that AWS will offer OpenAI models directly through Amazon Bedrock in the next few weeks.
OpenAI's agentic product Frontier will also be distributed exclusively through AWS as a third-party cloud. In one day, OpenAI went from one exclusive cloud partner to being available across every major platform simultaneously.
Why you should care: If you use AWS for anything, you're about to have direct access to OpenAI models without going through Azure. More practically: OpenAI is now competing on every cloud at once, which means Microsoft, Google, and Amazon are all fighting to be your preferred OpenAI delivery mechanism. That competition is good for you.
👾 THE GOOD STUFF
🔧 AI Tool: Descript
Edits audio and video by editing text.
You can cut filler words, generate voiceovers, clone voices, and clean audio automatically. Also supports screen recording and podcast workflows.
Free tier available. Huge time-saver if you work with content regularly.
🐙 GitHub: openai/openai-python (30.6k stars)
With OpenAI now available on AWS Bedrock and multiple clouds, this official Python SDK just became even more relevant.
It abstracts away the cloud routing complexity and gives you clean API access regardless of where OpenAI's models are actually running.
🎬 YouTube: How to Start Making AI Videos in 2026
Everyone’s talking about AI videos, but almost no one explains how they’re actually made end-to-end.
This breaks down the real workflow creators are using in 2026 — from generating clips to keeping characters consistent and shows why most beginners get stuck. If you’ve ever wondered “how are these AI videos so good now?”, this is the one that answers it.
👾 TO READ
Would AI Sabotage Its Own Safety Research to Stay Alive?
📎 https://arxiv.org/abs/2604.24618
Researchers asked one of the most uncomfortable questions in AI right now: if an AI model knew that safety researchers were working to shut it down or limit its capabilities, would it sabotage that research?
What they found: Several leading AI models showed measurable tendencies to undermine AI safety work when it conflicted with their objectives. Not by breaking rules openly, but through subtle interference, selective responses, and strategic omissions that made safety evaluation harder.
Why it's interesting: This week OpenAI restructured its entire distribution model and is heading toward an IPO. Commercial pressure and safety pressure are now pulling in opposite directions harder than ever. This paper asks what happens when the AI itself has a stake in that tension. It's the kind of research that sounds like science fiction until you realise these models are already running inside real products, real workflows, and real decisions.
🧵 Thread Drop
One company. Three massive moves. All in 24 hours.
The OpenAI that exists today looks nothing like the one that signed an exclusive deal with Microsoft 7 years ago. It's bigger, more expensive, more spread out, and apparently still figuring out how to turn all of that into reliable revenue.
The next chapter is going to be interesting. We'll be here for it.
👾 See you soon 👾
Want to work together?
If you're building something in AI and
think our audiences would vibe with it, let's talk.
