Skip to content

Podcast

Something Big Is Happening

The AI Daily Brief: Artificial Intelligence News and Analysis

Source ↗ ← All highlights
  • AI Built A Complete App By Itself
    • Nathaniel Whittemore describes a week where AI built, tested, and iterated an app autonomously for him.
    • He returned hours later to a finished product that he says was usually perfect on first test. Transcript: Nathaniel Whittemore In 2025, new techniques for building these models unlocked a much faster pace of progress. And then it got even faster, and then faster again. Each new model wasn’t just better than the last, it was better by a wide margin. And the time between new model releases was shorter. I was using AI more and more, going back and forth with it less and less, watching it handle things I used to think required my expertise. Then on February 5th, two major AI labs released new models on the same day. GPT-53 Codex from OpenAI and Opus 4.5 from Anthropic. And something clicked. Not like a light switch, more like the moment you realize the water has been rising around you and is now at your chest. I am no longer needed for the actual technical work of my job. I describe what I want built, in plain English, and it just appears. Not a rough draft I need to fix. The finished thing. I tell the AI what I want, walk away from my computer for four hours, and come back to find the work done. Done well. Done better than I would have done it myself, with no corrections needed. A couple of months ago, I was going back and forth with the AI, guiding it, making edits. Now I just describe the outcome and leave. Let me give you an example so you can understand what this actually looks like in practice. I’ll tell the AI, I want to build this app. Here’s what it should do. Here’s roughly what it should look like. Figure out the user flow, the design, all of it. And it does. It writes tens of thousands of lines of code. Then, and this is the part that would have been unthinkable a year ago, it opens the app itself. It clicks through the buttons. It tests the features. It uses the app the way a person would. If it doesn’t like how something looks or feels, it goes back and changes it on its own. It iterates like a developer would, fixing and refining until it’s satisfied. Only once it has decided the app meets its own standards does it come back to me and say, it’s ready for you to test, and when I test it, it’s usually perfect. I’m not exaggerating. This is what my Monday looked like this week. And here’s why this matters to you even if (Time 0:02:52)
  • Coding First Was A Strategic Lever
    • Making AI great at coding accelerates its ability to improve itself and unlock broader capabilities.
    • Nathaniel argues that coding-first progress was a strategic lever for self-improving models. Transcript: Nathaniel Whittemore If AI can write that code, it can help build the next version of itself, a smarter version which writes better code which builds an even smarter version. Making AI great at coding was a strategy that unlocks everything else. That’s why they did it first. My job started changing before yours not because they were targeting software engineers, it was just a side effect of where they chose to aim first. They’ve now done it, and they’re moving on to everything else. The experience that tech workers have had over the past year of watching AI go from helpful tool to does my job better than I do (Time 0:04:40)
  • Adopt Early And Practice Rapid Adaptation
    • Start using top-tier AI seriously and adopt early to gain a durable career advantage.
    • Build the habit of adapting and get comfortable being a beginner repeatedly as models change fast. Transcript: Nathaniel Whittemore Matt Matt writes, I’m not writing this to make you feel helpless. I’m writing this because I think the single biggest advantage you can have right now is simply being early. Early to understand it, early to use it, early to adopt. His advice then is, one, start using AI seriously and not just as a search engine. Basically, get the paid version, use the best model available, and use it for hard things. Second piece of advice, he says this might be the most important year of your career and work accordingly. Matt writes, the person who walks into a meeting and says I used AI to do this analysis in an hour instead of three days is going to be the most valuable person in the room, not eventually, Right now. Once everyone figures it out, the advantage disappears. Next, he says have no ego about it. The people who will struggle the most are ones who refuse to engage, the ones who dismiss it as a fad, who feel that AI diminishes their expertise, who assume their field is special and Immune. He has a bunch more, but then his final piece of advice is build the habit of adapting. He says this is maybe the most important one. The specific tools don’t matter as much as the muscle of learning new ones quickly. AI is going to keep changing and fast. The models that exist today will be obsolete in a year. The workflows people build now will need to be rebuilt. The people who come out of this well won’t be the ones who mastered one tool. They’ll be the ones who got comfortable with the pace of change itself. Make a habit of experimenting. (Time 0:06:35)
  • Software Bias Skews AI Expectations
    • Critics argue software’s structure makes engineers overestimate AI’s reach into less-structured fields.
    • Nathaniel notes many jobs involve unpredictability and human relationships that AI may struggle to replicate. Transcript: Nathaniel Whittemore Now, of course, 80 million people can’t look at a thing without getting some serious critiques. There is a healthy dose of personal invective aimed at Matt. There are dismissals via accusations of slop, basically saying that the ideas aren’t legitimate because they believe Matt used AI to write this 5,000 tome. Some people, I think, reasonably don’t love the COVID comparison, either because it feels too abstract, or it feels too aggressively doom and gloom, or because structurally, a virus That passes is different than a change that doesn’t change back. One of the most valuable critiques, I think, is the critique that basically says, other knowledge work problems outside of coding aren’t as instantly addressable and as easily addressable By AI as coding is. Isaac Saul writes, one thing I’ve noticed is that computer code is a really structured language and software is a defined problem space with a lot of defined patterns. So software people tend to think everything is a pattern and AI being really good at their job makes them overestimate how well it can do everything else. The truth is there is a lot more disorder on predictability and humanness in so much of our lives and our work that I don’t think AI applications will always or even often be able to account For. Matt, for instance, lists journalism as a job in trouble thanks to AI. Not that our industry needs more trouble. And it’s true that AI can read documents fast and do incredible research and even write clean copy and edit. It will probably eliminate or reduce the need for some jobs. But you know what it can’t do? It can’t work a source over for years on end. It can’t, doesn’t, and won’t bear witness to live events. It reminds me of the famous Goodwill hunting scene where Robin Williams is chastising Matt Damon about (Time 0:08:56)
  • Tool-Shaped Objects And Work-Shaped Outputs
    • ‘Tool-shaped objects’ describes AI systems that feel like tools but may produce outputs nobody values.
    • Nathaniel argues much modern knowledge work already creates work-shaped objects with limited real value. Transcript: Nathaniel Whittemore Now, one very highfalutin strand of this critique came from Will Manitas. He wrote another very widely viewed post called Tool Shape Objects. And for all the people ranting and raving about how good this one is, I think it basically uses good writing to trick you into thinking it’s made a point more profound than it actually Has. I think it actually secretly reveals something about the current state of work outside of AI entirely, which has some big implications as well. However, it was read enough that I think it’s worth excerpting as well. All right, friends, quick break to talk about a question I hear constantly. How do you actually move from AI experimentation to production without getting buried in infrastructure decisions? That’s where Rackspace AI Launchpad comes in. It’s a fully managed service designed to help enterprises build, test, and scale AI workloads through a guided phased approach. With AI Launchpad, Rackspace manages the infrastructure, GPUs, and core tooling, so teams can focus on validating use cases instead of building environments from scratch. You start with a proof of concept, move into a real pilot, and then scale into production on managed, enterprise-grade GPU infrastructure. Whether you’re testing inference at the edge, fine-tuning foundation models, or standing up a production pipeline, the goal is the same. Faster progress with less operational friction. If you’re ready to move beyond demos and actually put AI to work, take a look at Rackspace AI Launchpad and see how a managed (Time 0:11:27)
  • Favor Early Preparation Over Complacency
    • Prepare for asymmetric risk: underestimating AI is far costlier than modest overinvestment.
    • Prioritize early experimentation and learning to avoid professional obsolescence risks. Transcript: Nathaniel Whittemore I think Ethan is right, but let’s look at the implications of being wrong in each of the ways that Ethan suggests people are wrong. The implication for being wrong about the speed at which this AI diffuses across the workplace and society is perhaps overinvestment. It’s some extra time preparing when you could have used that time for other things. But ultimately, you weren’t wrong about the thing, you were wrong about the timescale. Now let’s talk about the implications of being wrong about fundamentally underestimating what AI can do and not preparing. It could literally mean, on an individual or an organizational level, professional extinction. Not that it will always be so, and I don’t think anyone can purport to know how exactly the lines between the AI haves and have-nots will shake out. It could be, and I hope it is the case, that there is plenty of time for everyone to catch up and adapt. That the skeptics of today, if indeed they are wrong, will have had time to be wrong and still adapt whatever it is that they do for work to the new reality without someone who wasn’t skeptical And embraced AI out-competing them. But I’m not sure that that’s going to be the case. The point, of course, is that the cost of underestimating AI is a hell of a lot higher than the cost of overestimating it. And so many people are just unwilling to change their priors. Now, one thing that gives me hope is that there are a lot of folks who are not AI people who are becoming more palatable messengers. (Time 0:20:35)
  • Seen Versus Unseen Shapes AI Impact
    • Bastiat’s ‘seen and unseen’ frames AI disruption: visible job loss vs. unseen new industries and opportunities.
    • Nathaniel highlights the unseen as harder to spot but critical to long-term outcomes. Transcript: Nathaniel Whittemore Conor Boyack wrote a follow-up about the seen and the unseen. It’s called AI Isn’t Coming for Your Future, Fear Is. He writes, I’m not going to argue that those articles are wrong about everything. AI is powerful. It is moving fast. The disruption is real and I take the concern seriously. But I’m going to tell you that the fear you’re feeling right now, that sinking sense that the rug is being pulled out from under you, is one of the oldest and most consistently wrong reactions In human history. It has a name, it has a pattern, and it has a track record of being spectacularly, almost comically incorrect. Not once or twice, like every single time. Conor writes that the single idea, written over 175 years ago, that is the master key to understanding every AI doomer headline you’ve ever read, is this. It’s from Frédéric Bastiat from 1850, when he wrote, There is only one difference between a bad economist and a good one. The bad economist confines himself to the visible effect. The good economist takes into account both the effect that can be seen and those effects that must be foreseen. Conor simplifies this to the seen and the unseen. He writes, He writes, The ones Bastiat said emerge only subsequently. These are the unseen. The new industries that don’t exist yet. The businesses that become possible only because costs have dropped. The creative work that gets unlocked when drudgery disappears. The entrepreneur who can now build alone what used to require a team of 20. The consumer who now has access to something that was previously unaffordable. The unseen is, by definition, harder to see. That’s the whole point. (Time 0:23:22)
  • Ask What AI Makes Possible Now
    • Ask what new things AI makes possible and reframe work around those opportunities.
    • Act with curiosity, not fear, to unlock creative and entrepreneurial upside from automation. Transcript: Nathaniel Whittemore They stare at the scene, extrapolate doom, and completely miss the explosion of new opportunity forming just outside their field of vision. Now, Conor goes on and gives lots and lots of evidence of this throughout history. He connects it back to AI, talking about how the scene effects are AI doing many of the tasks he used to spend hours on, while the unseen is being freed up to do higher-order work that he Never had time for before. The real risk, he argues, is not It’s AI. It’s mindset What can I do now that I couldn’t do before? That question, what is this making possible, is the most valuable question you can ask right now. About AI, about your career, about your life. The knitting machine didn’t ruin England. It made it the wealthiest nation on earth. The power loom didn’t destroy the textile industry. It expanded it beyond anyone’s imagination. The computer didn’t end employment. It created the modern economy. AI won’t shrink your future if you refuse to let fear shrink your vision. (Time 0:25:18)