New AI models from Google & Anthropic

Just when I thought AI was about to chill out for the first time, bombs where dropped about new models, capabilities and way too expensive subscriptions. The week started with OpenAI coming announcing that they are buying Jon Ives’ new startup IO for 6.5B. This is interesting also because the purchase was a stock-for-stock acquisition, which I presume means that the control of both the non-profit foundation and Microsoft have been diluted. The startup itself apparently did not have a name before the transaction, has no launched product, no users, but a huge price tag. The company is apparently creating a new physical device to interact with AI. SAMA at least appeared to be hyped about the new device.

But the AI news certainly did not stop there. Googles IO developer event also happened, and as you would guess, it was full of new AI hype. They launched their new Gemini 2.5 models, a new 250e Ultra tier AI subscription to use all their latest gadgets, new and apparently improved AI summary of google searches and about 100 other things.

And still the news did not stop there. Anthropic, not wanting to get outdone, launched Claude 4. From my preliminary testing and the launch post, the focus for Anthropic continues to be the tool use functionality and to be the best LLM for SW engineering.

- Roope Ripatti

Exciting news 

OpenAI acquires IO

As mentioned, this is a 6.5B deal to acquire a startup by Jon Ives. Seems a bit crazy in a good way.

100 things launched at Google IO

Claude 4 launch post

Upcoming stuff

Tue 27.5.

Wed 28.5.

Your weekly read

A bit of a random one, but I got interested in SAMAs past company control scheming. Super interesting story from early days of Reddit.

Last words if need be & something else?

Also, X was full of quality memes about Googles AI dominance after the IO event. This one was one of my personal favorites :D

-Roope