Brynjolfsson’s J-curve ratio: for every dollar on the model, enterprises spend up to ten on process, data, and change—why pilots stall and budgets lie.

A six-layer map of AI (infrastructure through application), fulcrum assets per layer, and how the stack forks into software AI vs physical AI.

How DeepMind picks a handful of transformative science bets per year, scaling laws in biology, and three kinds of intelligence reshaping research.

A developer take on why stronger AI coding tools can worsen maintenance, debt, and the gap between what ships and what stays understandable.

A strategic lens on AI distribution: commoditize complements to reinforce the model core. The piece compares Google’s historical playbook with Anthropic’s recent product moves, and argues startups can still win through focused specialization where giants cannot sustain depth.

A synthesis of new findings on AI-supported learning: shallow, scattered AI use can hurt outcomes, while strategic delegation of substantive tasks can free cognitive capacity for higher-order thinking. The key variable is not AI use itself, but instructional design that pairs offloading with verification, retrieval, and unassisted assessment.

What AI's biggest weakness reveals about our own minds.

This showed us that the activation of these patterns could actually drive Claude's behavior. And what our experiments suggest is that this Claude character has what we're calling "functional emotions," regardless of whether they're anything like human feelings. We look inside the model's "brain", the giant neural network that powers it, and by seeing which neurons "light up" in different situations, and how they're connected, we can start to understand how models think.

How did you evolve, how did you survive this technological adolescence without destroying yourself?” When I think about where humanity is now with AI,about what we’re on the cusp of,my mind keeps going back to that scene, because the question is so apt for our current situation, and I wish we had the aliens’ answer to guide us. There is a scene in the movie version of Carl Sagan’s book Contact where the main character, an astronomer who has detected the first radio signal from an alien civilization, is being considered for the role of humanity’s representative to meet the aliens. In my essa...

The tech entrepreneur on Claude Mythos, repercussions from the Pentagon dispute, and his message for the super-rich

Many of the implications of powerful AI are adversarial or dangerous, but at the end of it all, there has to be something we’re fighting for, some positive-sum outcome where everyone is better off, something to rally people to rise above their squabbles and confront the challenges ahead. In particular, I’ve made this choice out of a desire to: Yet despite all of the concerns above, I really do think it’s important to discuss what a good world with powerful AI could look like, while doing our best to avoid the above pitfalls. Because of this, people sometimes draw the conclusion that I’m a p...

We study empirical scaling laws for language model performance on the cross-entropy loss. The loss scales as a power-law with model size, dataset size, and the amount of compute used for training, with some trends spanning more than seven orders of magnitude. Other architectural details such as network width or depth have minimal effects within a wide range. Simple equations govern the dependence of overfitting on model/dataset size and the dependence of training speed on model size. These relationships allow us to determine the optimal allocation of a fixed compute budget. Larger models are significantly more sample-efficient, such that optimally compute-efficient training involves training very large models on a relatively modest amount of data and stopping significantly before convergence.

People often optimize for long-term goals in collaboration: A mentor or companion doesn't just answer questions, but also scaffolds learning, tracks progress, and prioritizes the other person's growth over immediate results. In contrast, current AI systems are fundamentally short-sighted collaborators, optimized for providing instant and complete responses, without ever saying no (unless for safety reasons). What are the consequences of this dynamic? Here, through a series of randomized controlled trials on human-AI interactions (N = 1,222), we provide causal evidence for two key consequences of AI assistance: reduced persistence and impairment of unassisted performance. Across a variety of tasks, including mathematical reasoning and reading comprehension, we find that although AI assistance improves performance in the short-term, people perform significantly worse without AI and are more likely to give up. Notably, these effects emerge after only brief interactions with AI (approximately 10 minutes). These findings are particularly concerning because persistence is foundational to skill acquisition and is one of the strongest predictors of long-term learning. We posit that persistence is reduced because AI conditions people to expect immediate answers, thereby denying them the experience of working through challenges on their own. These results suggest the need for AI model development to prioritize scaffolding long-term competence alongside immediate task completion.