Quite a few people have been asking me about the probability of doom and AI existential risk. Well, I am more worried about my X account getting hacked. And someone actually hacked it this morning. For some reason, it happens more often than I thought it would. So if you saw me posting some crazy Crypto ads, I am sorry.

New blog post: What's the Point of Computer Use Agents? There's been a lot of hype about CUAs recently. But where do CUAs shine? When should we use CUAs over API- or text-based agents? And what are the remaining research problems?

This is quite an elegant way to increase the probability of low-probability, high-reward trajectories. TailRL changes the credit assignment so that trajectories reaching reward levels that few other rollouts reach get more weight. Instead of only pulling up expected reward, it optimizes the probability of exceeding reward thresholds across the whole distribution, with a clean finite version of the gradient estimator and a nice best@k interpretation. It also reduces exactly to MaxRL when rewards are binary. The paper is very well written and has a ton of experiments across different domains. larger rollout budgets during training incorporate progressively higher order best@k , while the tail-focused objective increases the probability of sampling low prob but correct trajectories that become especially useful when you can sample many rollouts at inference.

Sharing lecture videos for the **How to AI (Almost) Anything/Multimodal AI** course I taught at MIT in spring 2026. This course became quite a hit the last time I shared it in spring 2025. Spring 2026's updated version contains updated topics on multimodal agents, reasoning, self-evolving AI, and new modalities like touch and smell. Also includes slides (not videos) of guest lectures on multimodal AI for health, design, manufacturing, cities, & transportation. Youtube playlist: Course website and materials: Today's AI can be applied to almost anything - from language to vision, audio, sensors, medical data, music, art, smell, and taste. This course covers the principles of AI (focusing on deep learning and foundation models), how we can apply AI to novel real-world data modalities, and multimodal AI that can process many modalities at once, such as connecting language and multimedia, music and art, sensing and actuation, and more.

Check out our new work on Tail-Likelihood Reinforcement Learning (TailRL), extending maximum-likelihood RL from binary to continuous rewards. Rather than optimizing only mean reward, TailRL maximizes the expected log of upper-tail probabilities, naturally placing more weight on rare, high-reward rollouts. Its gradient can also be interpreted as a mixture of Best-of-(k) gradients. TailRL requires only a simple modification to the advantage function, making it easy to integrate into existing RL pipelines. Across object localization, maze navigation, GUI grounding, and code optimization, TailRL effectively exploits rare high-reward samples and scales better with increased inference-time sampling. Check out a detailed thread by @stablegradients.

Happy to share that I have been selected as a Cosmos grantee for 2026 under the theme of Human Autonomy! @danish037 and I will be working on building an open-source AI detector. Thanks to @cosmos_inst for the support!

I fully agree with you Nando @NandoDF. But if governments fail to create the right conditions for AI talent to easily start companies (for a variety of silly reasons as you are pointing out), while those exact same researchers can easily build startups anywhere else, especially in the U.S., that's really really bad. AI talent is mobile, and countries that make entrepreneurship difficult will lose their top people. I know a number of researchers in the UK and Europe who are considering moving to the U.S.

Enterprise AI is in a wildly paradoxical state 🤔, and we’re reaching the inflection point that will resolve it. 💥 Enterprises want to differentiate with AI, yet rent the same intelligence as their competitors. Their workflows and expertise are highly specialized, yet they rely on generic models built to be good at everything. They worry about AI costs, yet pay premium prices for massive models where only a fraction (1%) of the intelligence is relevant to their task. 💸 And they demand control and sovereignty, yet rent the intelligence becoming core to their business. This is not a sustainable equilibrium. The next era of enterprise AI is specialized intelligence companies build, own, and compound. And we are at the inflection point of this transition. That’s the bet we made when we started @oumi_ai two years ago. Today we’re closing the loop: Oumi can now not only automatically build your specialized AI models, but also deploy them into production, learn from their production experience, and continuously improve them. The intelligence that your business runs on, becomes your differentiator. Your compounding advantage. The winners of the next AI era will turn their own data, expertise, and experience into specialized intelligence that nobody else can rent. Don’t rent your AI. Build it. Own it. Compound it.

Archer Aviation @flyarcher is building something big. Take a look at its latest technical blog, offering a first look at its efforts to build the world's first aviation foundation model. An incredible team to collaborate with.

Luna is one of my favorite models out right now, it also does really well on the computer-use benchmarks we tested On Odysseys ( and MyPCBench ( it scores 51% and 55.4%. To put this into perspective, it performs similarly to the heaviest models offered while being 20x+ cheaper. If you have the max codex plan, this is essentially unlimited usage. I am super happy about this level of model being offered at an even cheaper rate and would suggest try asking Luna to use computer-use in your workflows

I rarely do podcasts, as they tend to be less technical and more speculative, but I enjoyed this conversation with Forward Deployed @realbasilchatha. We talked about the early days of deep learning, what frontier AI labs are working on (data, infra, engineering), some of our recent research on agentic AI, prediction markets, and decision-making.

Looks like my X account has been hacked. Someone has been sending DMs from my account to random people, inviting them to schedule a meeting with me via a Calendly link. Those messages are not from me. Please ignore them. Help me retweet this message. Hey @X, I changed my password but I could really use your help resolving this.