Strong liability enforcement could be helpful in the AI debate. If your agent swarm goes rogue, you’re liable. If your weakly protected model gets jailbroken, you’re liable. If you serve a weakly protected OSS model, you’re liable.

Expectation: the age of the IDE is over Reality: we’re going to need a bigger IDE (imo). It just looks very different because humans now move upwards and program at a higher level - the basic unit of interest is not one file but one agent. It’s still programming.

AI diffusion is far more rate limited by having good evals than most realize. The kind of evals that you see for every model release are incredibly helpful, but only tell you the shape of general AI progress and the relative capability level of models. The far bigger space over time are evals on all the major workflows that enterprises do, down to the specifics of an individual company. This will be a huge space over time because you can’t automate what you can’t assess the progress on. Enterprises will not be able to go just on vibes.

The fact that open weights models are being discussed credibly at this level of capability should be a huge update for many. The implications of open models getting to frontier performance ensures that you can always have sovereign AI, have the ability to post train for your specific workflows, cost optimize for various workloads, and actually afford to do much more with AI (which opens up meaningfully different applications). Huge win for the applied AI layer.

Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

This works really well btw, at the end of your query ask your LLM to "structure your response as HTML", then view the generated file in your browser. I've also had some success asking the LLM to present its output as slideshows, etc. More generally, imo audio is the human-preferred input to AIs but vision (images/animations/video) is the preferred output from them. Around a ~third of our brains are a massively parallel processor dedicated to vision, it is the 10-lane superhighway of information into brain. As AI improves, I think we'll see a progression that takes advantage: 1) raw text (hard/effortful to read) 2) markdown (bold, italic, headings, tables, a bit easier on the eyes) <-- current default 3) HTML (still procedural with underlying code, but a lot more flexibility on the graphics, layout, even interactivity) <-- early but forming new good default ...4,5,6,... n) interactive neural videos/simulations Imo the extrapolation (though the technology doesn't exist just yet) ends in some kind of interactive videos generated directly by a diffusion neural net. Many open questions as to how exact/procedural "Software 1.0" artifacts (e.g. interactive simulations) may be woven together with neural artifacts (diffusion grids), but generally something in the direction of the recently viral xitter.cf/zan2434/status/2046982… There are also improvements necessary and pending at the input. Audio nor text nor video alone are not enough, e.g. I feel a need to point/gesture to things on the screen, similar to all the things you would do with a person physically next to you and your computer screen. TLDR The input/output mind meld between humans and AIs is ongoing and there is a lot of work to do and significant progress to be made, way before jumping all the way into neuralink-esque BCIs and all that. For what's worth exploring at the current stage, hot tip try ask for HTML.

Uber founder Travis Kalanick just inverted the entire automation panic. Everyone assumes AI eliminates human value. The physics say the opposite. Kalanick: “Let’s say the entire world, everything in our world, was automated except for plumbers. You had machines making buildings. You would basically have like a thousand buildings a day.” The algorithm can design a skyscraper in a millisecond. It cannot connect the pipes. When compute violently accelerates the speed of construction, the unautomated human becomes the ultimate bottleneck. And the bottleneck captures all the margin. Kalanick: “How valuable would those plumbers be? Extremely valuable. Those guys, each and every plumber would be like LeBron. Why? Because plumbing is the long pole in the tent to progress.” If the machine needs a human to finalize physical execution, that human doesn’t get replaced. Their economic value goes exponential. Kalanick: “You got so much efficiency everywhere else that you need millions of plumbers.” The market thinks automation drives human wages to zero. The physics dictate it drives the bottleneck’s wages to infinity. The next decade doesn’t belong to whoever out-computes the machine. It belongs to whoever stands at the exact point where the digital engine meets the physical world. Kalanick: “If we get to this place where autonomous cars are everywhere, if it was a thousand to one, you still probably have, I don’t know, 20 million jobs, 50 million jobs.” The panic over job destruction assumes a static volume of output. When output goes infinite, the system demands more human oversight. Not less. Waymo doesn’t delete the human. It shifts them from driver of one vehicle to director of a thousand. Kalanick: “Until we get super AGI, humans are valuable and they are going to become more and more valuable because they will be the long pole in the tent to progress.” You are no longer the engine. You are the grid.