US Government Builds AI Capacity from the Ground Up

0:00Welcome to the podcast. I'm your host, Jayden Schaefer. Today on the show, we're talking about a huge development coming out of Washington. I think this has a lot of big implications for technology and AI specifically. And this is how the U.S. government actually builds things in the modern era. So the Trump administration has announced plans to hire around a thousand specialists for a new federal quote unquote task force. Essentially, this is going to be a group that is going to work on AI systems, software modernization, and then also they're going to be doing like large scale finance and infrastructure projects.

0:31So there's a bunch of different interesting things that are kind of getting lumped into this. I think the concept, you know, of having a thousand specialists working in the government building things is the way that they're putting this together is kind of interesting. I think regardless of where you fall politically, it's kind of one of the most concrete steps we've seen towards the government taking technology execution seriously. And I think it raises some important questions about how innovation, talent, and also public service are going to intersect in the future, especially with what is unlocked by AI today.

1:02So before we get into the story, you probably pay for a bunch of different subscriptions to get access to all of the best AI tools. I paid for OpenAI, Grok, you know, 11 Labs for Audio, Gemini, and like a dozen other things. And it added up really fast. And so because of that, I built something called AIbox.ai. It's just$20 a month, and you get access to 40 of the top AI models all in one platform. So there's a bunch of text, image, audio, basically everything you need in one place. so you don't have to juggle a bunch of tabs, you don't have to waste money on overlapping subscriptions. If you want to try it out, you can click the link in the description to AIbox.ai.

1:41It's my own startup, and I'd love for you to let me know what you think about it. All right, let's get into the show. So the basic announcement is really straightforward. The federal government wants to recruit about 1 ,000 highly skilled technologists to work inside of government agencies on a temporary basis. The roles are expected to last about two years, right? So it feels like, you know, maybe they feel like we have this window where we need to make a big rush, or maybe it's for budgeting reasons. If they just say they're going to work for two years, then it's not like, you know, oh my gosh, the government's getting this much bigger.

2:13So they're going to focus on areas like AI, software engineering, data science, cybersecurity, technological project management. I think this isn't really about policy writing or theoretical research. The goal that they're working on is just to build systems and to modernize the infrastructure. They're going to ship software that works. This is obviously, I'm sure anyone that has visited a government website in the last five years knows the pain of the state of the government websites. I think for that alone, I'm excited. I think for a very long time, one of the biggest challenges in government was that technology implementation was never their strong suit.

2:51A lot of federal systems run on legacy code that is over 10 years old. So some agencies are still relying on programming languages and infrastructure that most modern engineers have never touched, which is really terrifying if you think about it. Because of this, there's a bunch of negative consequences. And so obviously projects are going to move a lot slower. The costs go up dramatically. The user experience is terrible. The UI on a lot of this stuff is horrible. I think government employees and the public are both frustrated by this. And so historically, the solution has been to rely heavily on contractors, right?

3:28Outside contractors. And while there are some great contractors out there, I think the model has also led to bloated costs. Like it's no secret or surprise that when the government pays for something, they pay 10 times what they should. The timelines are very slow. I mean, we've all heard the horrendous stories of, you know, Oracle charging like New York City$600 million to create like a records keeping database for, you know, different projects. And it's just, you know, it's eye watering and feels pretty gross. There's definitely a massive lack of internal technical understanding of the government.

4:00So then they just get charged crazy amounts of money. And is there corruption or other things involved? I mean, I won't speculate on that, but it's not pretty usually. So in any case, when the expertise lives entirely outside the organization, I think it's really hard to make long-term, like good long-term decisions. So hopefully, you know, this is what is going to be solved here. Will it happen? That is definitely left to be seen. The jury's out on that. But the tech force is trying to approach this in a little bit of a different way. They're going to try to bring the experts inside, not just have them as contractors.

4:32and then they're going to work directly within the agencies inside of teams at the agency. So this the goal is they're going to be helping with design, building, and then deploying systems alongside career civil servants. So people have always been working in the government. I think the idea is not just to deliver projects, but to transfer knowledge and then also improve internal capabilities. So it's like, honestly, in a way, it's like we're upskilling a lot of federal workers who might have been working on legacy systems. Sometimes it just takes a person with a fresh set of eyes to come in and look at the system to say, okay, this is how we need to improve and change it.

5:04I think another important detail is the temporary nature of this. This isn't, you know, people asking to commit their entire years to government service. It's more just like a fixed term fellowship. You know, some people call it a tour of duty. But basically, you come in, you work on some sort of high impact problem. And then you decide whether to stay or then go return to the private sector. I think that structure matters because it kind of changes who might apply for the job. A lot of talented, you know, developers or technologists, they're open to public service, but they don't really want to commit to like a long-term thing, especially if it means lower pay or slower career progression.

5:43So a two-year commitment with really competitive compensation, it's going to lower the barrier significantly. I think it makes public service feel more accessible and also more aligned with modern career paths. And then speaking of compensation, a lot of the reports, early reports that are coming out about this are saying that the salaries are going to be competitive with mid to senior level private sector roles. So not necessarily, you know, the top of the market Silicon Valley compensation. I don't think the government's trying to compete with Mark Zuckerberg's, you know, billion dollar offers to the top AI researchers.

6:12But I think it is enough to attract experience like professionals, especially those who are motivated, you know, maybe they're motivated by like impact or scale rather than just how much money they make. So I think the types of projects that the tech force is going to be working on are they have a few different focus areas. First, AI data systems. So this includes building AI tools to improve efficiency, automating routine processes, detecting fraud, they'll be analyzing large data sets, they'll be, you know, it'll help to support better decision making across agencies. I think AI is particularly good for, you know, it's going to be well suited for government use cases where there's like a lot of structured data and repetitive workflows.

6:55The next thing, like I mentioned, is financial systems. They're going to be doing just general IT modernization. There's a lot to watch out for on this deal, how it's executed, how they're going to be doing the recruiting, if there's going to be scope creep. I mean, just are they actually going to be able to get anything done with the bureaucracy and red tape that is the federal government? So we'll be following up on all of that and keeping you informed. But thank you so much for tuning in to the podcast today. I hope that this is your number one spot for the latest in AI and business news. If you enjoyed the episode, it would mean the world to me if you left a rating and review on Apple podcasts or on Spotify or wherever you get your podcasts.

7:33And also don't forget to check out AI box.ai. We have an incredible playground there with access to the top 40 AI models all for 20 bucks a month. So so link in the description, you can check it out AI box.ai.