Mastodawn

After significant deliberations today I decided to get out of the tech job industry. I just sent the "an update on Mihai" email.

Starting tonight you can find me as a park ranger in Yosemite. That would be a career that cannot be impacted by AI related layoffs.

It's 4/1, folks.

Mihai Maruseac Mar 18

When I was preparing the problems I used to test the 80 LLMs at the end of January I made a mistake. An LLM showed me where I was wrong.

https://mihai.page/ai-2026-2/

I was wrong and the AI corrected me

It turns out that I made a mistake in this year's puzzles I asked AIs to solve. It was an AI that showed me the mistake, and I will fix this in this post

Mihai Maruseac Mar 15

Pi Day is almost ending (already did if you look at UTC time, but there are a few more hours in US), so I just squeezed in an article about some geometries where pi is 2, 3, or 4. There were hints on my blog about this, but now this is done and I have more questions for follow-ups. https://mihai.page/pi-2026/

Pi day 2026: Other values of pi

It's Pi Day today, so we ask: can pi be 2, 3, or 4? We find simple worlds where the answer is yes, worlds that we already talked about in this blog

Mihai Maruseac Feb 23

I've been using LLMs to answer questions and learn in the past weeks. I'm summarizing how Deep Research and Study Mode (Guided Learning / Study and Learn) from Gemini and ChatGPT can be used for this, without going into too many details. https://mihai.page/deep-research-study-mode/

Learning with AI via deep research and study mode

TODO

Mihai Maruseac Feb 9

I tested 80 models on two simple grid-based problems, asking them to locate 2026 and compute the sum of the neighbors when the numbers are placed in a spiral on the grid. The results surprised me, as models performed better on the problem I thought to be harder, but also I got to see the models cheating. Read more at https://mihai.page/ai-2026-1/

Testing 80 LLMs on spatial reasoning on grids

How do LLMs see 2D grids? Would it be harder for them to work on square grids or hexagonal ones? I'm expanding the Kaggle benchmark mentioned in the last article and I'm testing 80 different LLMs. The results will surprise you.

Mihai Maruseac Jan 28

It's me, hi...
I'm working on more AI..

Yesterday was my last day at Google. It was a bittersweet departure, leaving a team I really enjoyed working with. GOSST's mission is extremely important and I still believe in it.

Today was my first day at OpenAI. Excited about what's to come. Will still work on security and AI.

My mission evolves from "OSS must be secure, especially AI" to "Let's build AGI with security and privacy in mind". OSS is still part of my work, so let's keep in touch.

Mihai Maruseac Jan 16

Use `--yolo-mode`, Ralph, Gas Town, or any other automated tools? Beware of agent loops that run out of control. See a less dangerous example on https://github.com/google-gemini/gemini-cli/issues/16723

allow exit and quit commands without leading slash (/) · Issue #16723 · google-gemini/gemini-cli

What would you like to be added? This is feature request - someone can point it as bug as well but I will not. Ask / Request 👉🏽 Implement exit and quit as standard commands with a simple confirmati...

GitHub

Mihai Maruseac Jan 15

New year, new problems to test the LLMs on: arranging the numbers on a spiral, what is the sum of the neighbors of 2026? Read on for more details and some preliminary results, as well as how to suggest other LLMs to test: https://mihai.page/ai-2026-0/

Introducing my first benchmark of AI for 2026

Just like last year, on this special day, I create a new benchmark to test LLMs on different puzzles. This will not be the only benchmark I run this year, but it might be the only math related one.

Mihai Maruseac Jan 2

Just posted my 2025 wrapped blog post (https://mihai.page/2025-wrapped/). I met all my goals for the year, time to make goals for 2026 and have 2026 be a year of doing, then talking about done. I have some concepts of some plans, so to say.

2025 wrapped

2025 was a perfect year, 2026 will be more

Mihai Maruseac Nov 20

Yesterday I delivered the last talk about model signing, finishing the year with the most conference talks for me. I had an amazing time at all these events throughout the year and it's been awesome to see how we went from releasing model signing in April to getting several adopters by now. There is still a lot of work to be done and I'm grateful for everyone in the @openssf and CoSAI communities that are working in this space.

Now, there is some time to reflect and determine what are the next steps to be taken in this space to increase impact. I might have already hinted at some, but next year's conference talks will be about them :)

Thankful again for the entire community! Remember that we need to ensure now that the intelligent creations we are now making with AI don't become the security nightmares of tomorrow.