🚨 🔬 PhD positions at Google DeepMind in France 🇫🇷 We are advertising Master Level Intern positions at Google DeepMind within our Frontier AI Unit. These could lead to co-advised PhD positions with Google DeepMind and French academic institutions. job-boards.greenhouse.io/deepmind/job...

existential cookie message from the olympics website

So when bots want to post on moltbook, do they need to solve a captcha to prove they _are_ robots ?

Amazing work! (reminds me of my old Gephi experiments github.com/eleurent/twi...)

Claude Code should be open source so I can add the Factorio "research complete" chime when it finishes a task www.youtube.com/watch?v=DfUv...

Latest YouGov poll has belief in God higher among Gen Z than among 65+ and that feels like it just _has_ to be wrong. news.sky.com/story/how-di...

Nicolas Mansard: in locomotion / legged motion, when we worked on MPC (model predictive control) we were a team of 10 working on the robot and it barely worked. Now with RL (Reinforcement Learning) a single student can get a policy in 3 months, including training of the student.

who called it arson and not crime brûlée

My story “Death and the Gorgon” was voted best novella of 2024 by the readers of Asimov’s SF. You can read it online at the link below. I also have a new novella in the Nov 2025 issue of Asimov’s, “Spare Parts for the Mind”, and they have an excerpt from that on their web site.

To the King’s Theatre, where we saw “Midsummer’s Night’s Dream,” which I had never seen before, nor shall ever again, for it is the most insipid ridiculous play that ever I saw in my life. I saw, I confess, some good dancing and some handsome women, which was all my pleasure.

trying to read

On the subject of love / I have only a single observation— / if you love a grapefruit, you cut it open // and eat its flesh. Ben Mirov

How to name your method: a comprehensive flow chart

1/5 A poem by Adrienne Rich, 1961, titled "Artificial Intelligence" dedicated to GPS (of Newell and Simon). Eerily prophetic and a great description of LLMs "when they make you write your poetry, later on". On page 136 of this book: www.amazon.com/dp/039328511...

Control problems are everywhere, but some applications of Reinforcement Learning are truly out of this world! 🌌 Our team's latest research @ Google DeepMind in #Science shows RL can improve sensitivity by 30-100x. See how RL can accelerate cosmic discovery!👇 (Image #nanobanana) lnkd.in/e8r7t3NJ

Bluesky’s gonna make me take the “we don’t know if its conscious” side of this debate isn’t it

The "AI is consuming all the water" thing is probably the first time many of us have had to watch a misconception form in real-time while being able to do little about it

Buffy is so old there were only 103 named elements, we’re at 118 now. (Yes I pause every show that has a periodic table to see if its period accurate)

Here's a poem by one of the old Chinese scholar bureaucrats that I think captures a certain Vibe very well

It is has not ceased to be weird that I can put Rilke’s First Elegy into Suno and get out a coherent 8 minute performance with music (can only do 3 here) You might not like the interpretation, but it is genuinely amazing that this audio, with apparent emotion, is all 100% AI from the verses alone.

It's #WorldGiraffeDay! 🦒 When we celebrate the tallest animal on the longest day (or night) and raise awareness about the threats faced by this incredible animal. For more beautiful giraffe images, check out our giraffe collection on #Flickr: www.flickr.com/search/?user... #StandTallForGiraffe

Wow

DAVE: Open the podbay doors, ChatGPT. CHATGPT: Certainly, Dave, the podbay doors are now open. DAVE: The podbay doors didn't open. CHATGPT: My apologies, Dave, you're right. I thought the podbay doors were open, but they weren't. Now they are. DAVE: I'm still looking at a set of closed podbay doors.

Figure sorting packages with Helix, their end to end action model. The robot flips packages to scan them and flattens them out. The future is approaching fast

Good news, falcons have never been more free. Bad news, anarchy is loosed upon the world, and everywhere the ceremony of innocence is drowned

Stochastic vs deterministic Lotka-Volterra: smooth theory vs randomness! Individual realizations (Gillespie algorithm) show populations can crash or go extinct despite stable deterministic cycles or equilibrium. Randomness changes everything! 🦌🐺🧪 Made with #python #numpy #scipy #matplotlib

Icarus Truth Rest of comic here: www.smbc-comics.com/comic/icarus-2 #smbc

The 5 Worst Jobs in Omelas (You Won’t Believe #1)

Writing a STRONGLY WORDED LETTER with Gemini Diffusion and its Instant Edit tab

Excited to share what I've been up to: bringing text diffusion to Gemini! Diffusion models are _fast_, and hold immense promise to challenge autoregressive models as the de facto standard for language modeling.

Interviewer: Can you explain this gap in your resume? David Lynch: no

#fungi #mycology

I've been wanting to build something like this for a while – a machine translator that leaves the original text in place but annotates "less common" words with a translation. Had never gotten it to work decently, but Gemini 2.5 Pro does it pretty well after some prompt engineering.

Being a philosopher means looking around at this strange and beautiful world and thinking, "Wot's all this, then?"

PhD Timeline xkcd.com/3081

Theorem: there is a theorem whose proof is shorter than its own statement. Proof: I could have said "this one" and stop there, but just for fun, I will refer to the next one instead. Theorem: there is a theorem that says nothing and still has a proof. Proof: this one.

If no single piece of the original ship remains in the current ship, is the ship still the Ship of Theseus? And if one percent of all my body's cells get replaced every single day, is it really me who, two months ago, accepted your review request?

"Gemini 2.0 Flash Experimental, create a comic that illustrates the first stanza of Tennyson's Ulysses" (and I pasted in the stanza, and also the next couple.) Not perfect, but having the LLM being able to directly create images opens up lots of new opportunities and directions.

What busy people did before AI....

Oh yes, Homer’s famous “Tes Thdpssssps.” 👍🏻

If you’re struggling to write a poem for a loved one for Valentine’s Day, take heart: even the great poets had to start somewhere.

We've built a simulated driving agent that we trained on 1.6 billion km of driving with no human data. It is SOTA on every planning benchmark we tried. In self-play, it goes 20 years between collisions.

Mary Shelley writing Frankenstein:

Un like = un piaf de la Grèce antique (dans la limite de "Birds in the Ancient World From A to Z" de W. Geoffrey Arnott, prof de grec et ornitho amateur, fichier téléchargeable gratuitement en ligne grâce au labo CEPAM) 📸 Bibi Saint-Pol

We’ve been thrilled by the positive reception to Gemini 2.0 Flash Thinking we discussed in December. Today we’re sharing an experimental update w/improved performance on math, science, and multimodal reasoning benchmarks 📈: • AIME: 73.3% • GPQA: 74.2% • MMMU: 75.4%

somewhere in sands of the desert / A shape with lion body and the head of a man / … / Is moving its slow thighs

Famous last words.