This Week in AI for Ridiculously Busy People

0:00Today on the AI Daily Brief, this week in AI for ridiculously busy people. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. All right, friends, doing a quick experiment here. The AI Daily Brief is obviously quite an information-dense podcast. Despite curating the whole world of AI things happening, it can still be a pretty high barrier to climb for people who are paying attention more casually or just don't have time to dedicate 20 or 25 minutes a day for AI news. So for those of you who are looking for something that's closer to five minutes to send your colleagues who need to know exactly what was going on in AI that week.

0:36That's what this is for. Let me know how you like it. First up, let's talk about the biggest theme of the week, which was absolutely by far token efficiency. I made the argument on Twitter that every AI company is now in some way, shape or form a token efficiency company. We have moved officially from the token subsidy era, where the per seat models of companies like OpenAI and Anthropic were allowing people to consume thousands of dollars worth of AI tokens for tens or hundreds of dollars. now we're in the token shortage era, where all the business models are moving to usage-based models, and everyone is having to adapt.

1:07This week, that adaptation looked like Uber putting$1 ,500 monthly limits on employees' AI usage, and Walmart having to cap usage of their tool for it being too high in demand, and comments from companies like TSMC suggesting that this shortage is not a short-term thing, but is going to last years. Importantly, though, the market is responding. AI software engineering agent company Factory introduced native model routing that can figure out what the right model is for a task, including models that are cheaper or not state-of-the-art, which they say can maintain state-of-the-art performance while cutting cost by a quarter.

1:39Perplexity introduced a new system that combines a hybrid, local, and cloud-based inference system, which has benefits for both costs and privacy. Harvey announced that it had collaborated with Fireworks.ai to build a worker advisor agent where an open-weight worker can delegate complex tasks to a closed-source frontier advisor powered by one of the state-of-the-art models and found that it outperformed the state-of-the-art model alone on the legal tasks for just a fraction of the costs. Microsoft, meanwhile, is clearly trying to bring this sort of capability to the rest of the market, saying that when they collaborated with McKinsey to post-train a model on McKinsey tasks, it beat GPT 5.5 performance at a tenth of the cost.

2:19TLDR, the token shortage is here, but the market is responding. In terms of what you should be playing with this weekend, it is absolutely Codex updates. Codex announced an expansion of their plugin ecosystem, new annotations, and a new feature called Sites. Annotations allow you to edit and interact with specific parts of any given website or document that you're working on. Plugins are expanding to include functional-based plugins. So for example, you can use a plugin pack for salespeople that comes with connectors to common tools and skills that are relevant for that function. And Sites is maybe the most interesting one where you can turn anything you're working on inside of Codex into a website or web app with a single click, which I think will help make websites a fundamental unit of knowledge work in a way that they're not right now.

3:00Now, right now, that's just available to business and enterprise users, but if you have access to it, you should definitely go play around with sites. Next up, the thing you have to pay attention to, even if you don't think you care, is the question of who owns AI. The stakes in the policy discussion are getting higher as the model capability increases and we get towards IPOs. Bernie Sanders came out with an op-ed in the New York Times this week, suggesting that the government own a full half of the major AI labs. And while they might not be getting 50%, it seems like the Trump White House is considering taking equity stakes in the big labs, meaning that the Overton window on this sort of government company collaboration is changing very quickly.

3:36We also got policy-related papers from both Anthropic and OpenAI this week, talking about how they were seeing the early signs of recursive self-improvement in today's AI systems. So I think the policy discourse is about to get much louder very soon. The biggest takeaway for enterprises, simply put, you are now in the token efficiency business and you need to be thinking about that A, architecturally, in other words, things like model routing and model selection, context management, and making sure people aren't wasting cycles trying to find the right context, but also B, in supporting best practices and specifically training.

4:09The cost of the average enterprise's failure to train their people on these new systems has never been higher. If you don't have a company-wide agent-centric training program yet, you are officially behind. Biggest takeaway for solo practitioners, meanwhile, is that even if you are proficient with agents right now, this is a good time to start building your systems because the cost equation is going to be more of a challenge. That means a smaller version of things like context management, like integrating skills and more, TLDR now is the time to build systems. In terms of what to watch for next week, well, of course, the big thing will be the SpaceX IPO, which will be the biggest IPO in history.

4:48And tell us a lot about how the market is thinking about these companies. So there you have it, the week in AI for ridiculously busy people. Let me know what you think about this format, and maybe it'll become something that we do more often. Appreciate you listening or watching as always. And until next time, peace.