Research Scientist @ Google DeepMind. Formerly Robotics, now AI Safety. Has a blog. Views are my own. "Adversarially disengaging Twitter profile"
Joined December 2012
- Tweets354
- Following31
- Followers2.8K
- Likes30
Pinned Tweet
I'm on Bluesky now. I plan to cross-post blog posts to both platforms for the time being, we'll see about the other stuff.
bsky.app/profile/alexirpan.b…
Two years ago I wrote a post about my decision to work on AI safety & alignment. In it I guessed 2% chance of doom.
Since then my vibe has gotten about 2x worse (4%). It doesn't feel like safety's improved at the rate needed to keep up with capabilities. It's not looking good.
I'm working in AI safety now.
alexirpan.com/2024/08/06/swi…
I plan to ignore the controversy on whatever's going on with Navier-Stokes. However, I also noticed my reaction was "huh, okay" rather than "wtf??????" at an AI solving a Millennium problem. (I expect the result to stand up, controversy aside.)
I'm the frog, it's me, I'm boiled.
Today is my 11 year blogging anniversary.
alexirpan.com/2026/08/18/ele…
Alex Irpan retweeted
Man I love being right
The GDM AGI safety team is hiring! If you have concerns about the implications of increasingly powerful AI, you should apply. Things are kinda wild right now and I don't expect the wildness to slow down.
In retrospect "AI CEOs become celebrities and attract corresponding levels of desired and undesired attention" was a very obvious prediction that many people failed to account for (including me)
Inspired by talking to a few too many optimists.
alexirpan.com/2026/05/17/ai-…
1. Obviously terrible to have a Molotov thrown against your house, not appropriate response
2. Of all analogies to make, "ring of power" is a choice, given the story's theme that the only way to stop the ring's destructive power is to destroy it.
blog.samaltman.com/2279512
There is now another amicus brief filed by a number of former high ranking military officials (up to Admiral level), arguing these actions hurt the military's adherence to the rule of law.
storage.courtlistener.com/re…
You know, when I switched into safety, I was a little worried it was too early.
Between the decline of coding by hand, OpenClaw YOLOing, increasingly eval aware models, and DoD pressure to let AI be used for surveillance and autonomous weapons
yeah
It wasn't early
Here's my MIT Mystery Hunt post for the year
alexirpan.com/2026/01/29/mh-…
I didn't know where this post was going when I started and I'm not sure where it went now that it ended, but that felt correct in some way.
alexirpan.com/2025/11/16/aut…
First paper since switching into AI safety team🎉
We look at problems that could be solved if the model behaved consistently over a set of prompts, and tried training that in output space and internal activations. Both were effective. See thread or paper for details.
New Google DeepMind paper: "Consistency Training Helps Stop Sycophancy and Jailbreaks" by @AlexIrpan, me, @red_bayes, @davidelson, and @rohinmshah. (thread)
ALT The abstract of the consistency training paper.
> switch to AI safety
> no safety papers to cite in reviewer profile
> only get assigned robotics papers
Apologies in advance as I try to crash course the past year in a few weeks...
Today is my 10 year blogging anniversary
alexirpan.com/2025/08/18/ten…
For the past month I have been working on a blog post about niche MLP fandom drama. Well here it is.
alexirpan.com/2025/07/21/bab…
