The Big Deal About the Big AI Announcement - What is next, and what does that mean to competition, IT, and the board of directors at your firm

So the $500B AI ‘Infrastructure’ announcement at the White House this week is a bigger deal than you think, with several ripples and problems that most folks don’t see.

This article includes a call to action for every enterprise, some good news, and a view of what should happen at the board level.

It is the most difficult time for Boards of Directors in twenty years maybe more. I have some suggestions on key board actions/questions.

We will walk through the true scale of this and why it is more significant and more important than it appears on the surface.

Scale

$500B on tech is, well, a lot. It’s more than you think. Let me give you some perspective and some support for change.

$500B would let me buy all the residential real estate in Las Vegas, Detroit, Charlotte, Jacksonville and half of Columbus, Ohio.

$500B is more than the annual GDP of Thailand or Austria…the entire nation.

$500B is enough to buy Ford, DELL, GM, UBER, Marriott, Micron Tech, Chipotle, Kroger, Southwest Airlines, Occidental Petroleum, eBay, Moderna, EA, and Lyft.

Enterprise Technology Scale

Here are two enterprise examples: Before it was broken up, the original Motorola spent $1.4B on IT a year. We cleaned that up and doubled output while reducing spending to $865. At massive Aviva in the UK, we reduced IT spending from $2.2B to $1.6B and tripled output. More recently, I have had a detailed view of three giant companies: IT in retail, FS, and Tech, and the scale/situation is similar.

So that means two things:

I. The $500B they want to spend is over four years so that’s $125B a year which is 5000% of the sum of the entire IT spend for two enormous F100 companies.

II. Most IT shops have tech debt, duplication, process/system inefficiency, low-value spending, bad deals, and leadership upsides. So, during a time of great tech change or opportunity, building business cases for upgrades or changes over a three-year period can be funded by real impact.

In all fairness, the CIO job is uniquely challenging.

Imagine working in a construction field where the base construction material had 30+ tectonic shifts/developments, none coordinated with each other, in three decades. Layer on top of the enemies that are trying to break your construction and steal your value (hackers). Layer on top of that a boss and board who usually don’t really know what you do, let alone where your field is going. Imagine a job where punishment is swift for failure, which makes CIOs play defense, sandbag plans, overuse external resources, and keep innovation small and safe.

$500B is Big But It’s Bigger Than That in Three Ways

I.  More Money

The $500B was three firms plus a UAE sovereign fund investment arm. The rest of the top ten tech behemoths are also going to pour money into this. There are other emerging companies getting billions in investment. So, in the next years its easily a trillion dollars.

II.  Government Afterburners

Favorable, positive, engaged government support stretches those dollars in several ways, so from an impact perspective, it is still bigger. This is a good thing, a very good thing. It is a key to economic competition and national well-being.

III.  Technology Evolves Differently Than Most Things

This is infrastructure-level technology. The entire history of that is smart people rearchitect, redesign, and restructure how this fits together and how it works to make it faster, better, and cheaper. New things get invented all the time and integrated into the creation and function of tech, especially infrastructure-level stuff. The clever people at Ocient rearchitected the infrastructure stack for hyperscale data, for massive data analysis, in some cases it is 1000x more efficient. Yep, 1000x. Vantiq takes four layers of infrastructure headaches and combines them into a platform that not only takes out cost for IT but massively reduces the amount of work for impact. API platforms, Middleware, AppServers, testing platforms, and a mountain of stuff at Amazon Web Services are all examples. Complex, expensive things in tech get turned into Legos that can be taken out of the box and plugged in. That happens all the way down to chips with smart innovations and architecture-level design, re: Moore’s Law.

DeepSeek V-3 is an LLM (Large Language Model) out of China. Think of LLMs as a key part of the base infrastructure for GenAI and Agentic AI.  It was recently released and is an innovation at the architecture and design level. It is 200x cheaper than ChatGPT4 (for token input and output ). DeepSeek R1 (a version designed for complex reasoning) may be 400x. The architecture innovation allows DeepSeek to use about 11x less compute. Henry Ford built a production line that was 10x more labor efficient than the competition. 400x is me jogging versus an F-35 at top speed or a giant tortoise versus a cheetah. A 200x improvement is 15 years of Moore’s law. We just saw that in one step.

The point here is that the entire history of tech, architecture, and design innovation radically improves these things. Even given that history of the tech world these are improving particularly quickly. So the $500B over four years feels like $2T in 2025 spend.

Meanwhile, Back in the Real World

The biggest impacts normally come on top of base layer infrastructure innovation that creates myriad growth ripples. Think of steam power, electricity, the telegraph, railroads, the Croton Aqueduct (spawned modern NYC), the internet,  and more recently 4G. Nothing super exciting happened with Telco’s introduction of 4G, but it allowed several big, new, disruptive things to be built on top of it—things like Facebook, Uber, Spotify, TikTok, Netflix supercharged Cloud Services, and eCommerce. Rockefeller’s Standard Oil and early giant success SEARS were built on top of the leading-edge networks of that era: the telegraph and the railroads.

So, we will see the Lego-ization of difficult tech stuff coupled with new, improved offerings and models on top of the LLM world. Clearly, the entire coding world will change. The nearshore/offshoring industry is about to undergo a tectonic shift.

But in the real everyday world for the Enterprise the CIO still struggles with cybersecurity, data (governance, integration, architecture, leverage), tech debt, ,difficult construction and a mountain of maintenance. Boards will start to realize that tech is the operating system for future companies, and the competitive race looks like it’s not just giant tortoise speed but also cheetahs.

A Real Current Example

One example of next-gen tech that fits the above story and that I am particularly fond of is VANTIQ . I am doubly fond of it because I am on the board, am an advisor, and am personally invested (monetarily). It has taken the data integration headache and internalized all that complexity so the IT folks don’t have to. It has layered on AI, GenAI and the capability to orchestrate AI (the key to Agentic AI) similarly. It has Lego-ized it. It has then put in a low code - no code, easy construction space to derisk again and accelerate the CIO headache of responding agilely to the business. The icing on that cake is it can also do all of that in real time because it is architected differently.

Real time is where the real world happens, but it is also a portal into the future. If I can see patterns and indicators in real time I can intercept bad outcomes not in an emergency code blue moment but a week a day or an hour in advance and manage resource flows proactively. Better, proactive resource flow management is key to everything from retail/inventory management, healthcare, smart cities/spaces, energy plants, and even warfighting . This is seeing over the horizon in an intelligent, information-based way. Data is nice but what you really want is a chain that turns data into information into intelligence into timely action. Then, you want a construction layer to instantiate and make permanent the changes you created from your new, more intelligent viewpoint.

Yes, it’s crazy hard to build that, and it took years by some of the best, but it is precisely the continuation of tech’s historical trend. It will improve outcomes, IT agility, and operational quality while lowering the cost structure in the business and IT. It is an example of building on top of infrastructure changes.

What Does All This Mean for the Board

The functional relationship between boards, CEO and their IT groups has been generally bad since CIOs were invented in the 90s in response to big tech companies having a sales model that felt more like a Viking invasion than a service relationship.

I have seen several surveys in which 70%+ of CEOs think their IT group is bad. Korn Ferry reported that while the average tenure for CIOs has grown by 50% over the last decade, it is still half that of the average CEO tenure. I recently had a board member at a huge European-based global firm refer to their CIO as a head of wet lettuce. I am pretty sure that is a bad thing. However, as I said above, the CIO job is tough.

Boards have to have a strategy refresh moment now. Think of this as warfare, and we have just discovered gunpowder or guns.

When Henry Ford first had his production line in 1912, there were 400 vibrant companies in the US. Within 20 years, there were 40, and three had 80% of the market. 1912 would have been a good year for a strategy refresh moment at the board. At today’s velocity, those 20 years are now six or eight.

Discussing where a firm should be doing GenAI and how it is insufficient. Bravely letting almost everyone have GenAI, including the ability to build tech solutions is a monumental mistake. It is something I have seen advocated in several top business publications. That misses the big picture and will create a layer of techno spaghetti at your firm that will be nearly impossible to unravel.

The speed and intensity of competition are accelerating, while the LOWP (the Land Of What’s Possible) with tech is exploding. The structure, focus, and capability map for Enterprise IT are changing. A board can ask four quick, key diagnostic questions to gauge how prepared their IT shops are for the near future.

(Anna Catalano and I are writing a book on this anchored in innovation married to pragmatism).

This is the framing the board needs to understand. Here are two simple charts from an interactive lecture  I did in Nov. 2021 at the marvelous Naval Post Graduate School in Monterrey, Ca in the Applied Design for Innovation program where they still let me give talks. It was anchored in explaining the 4th Industrial Revolution that came out of Davos and the underlying tech thinking. The first chart is what those technology's paths looked like looking back from 2021. The second chart is what I said they will look like just four years into the future, i.e 2025. I have added a new third chart looking back from 2030. This is what boards have to contend with. Note the vertical axis and what it measures.

The same two initial graphs were in a chapter I did in a book on technology and Special Ops Forces in Jan. 2021, published by the Center for Global Security Research at Lawrence Livermore National Lab and Special Operations Command.

Conclusion

This is clearly a difficult time and like all creative destruction cycles a lot of folks end up on the destruction side.

More importantly, this is a time of great opportunity.

Our prosperity and national competitiveness are tied to how well we compete on the global economic stage. These investments and this trajectory bode well for us. But again, a lot of the impact and value come above the infrastructure layer. No telco got rich rolling out 4G, but huge value was created on top of 4G. There are just a handful of patents on Velcro that made a lot of money, and there are a thousand patents on clever uses of Velcro that made more.

These modern technologies, especially at the enterprise level, on top of the infrastructure, are a lot like rugby. Once it’s been explained to you, it is actually conceptually pretty simple, although clearly very skill-oriented and full of nuance. The big thing is you do not become great at rugby by studying it or watching it. You have to get out on the pitch and play. You must be okay with bumps, bruises, failure, and surprises that come with that learning curve. This is a hands-on tech era. As we said above, the good news is tech firms are continually working to Lego-ize the tricky bits and internalize the complexity, making adoption an easier win. Uber under the covers is 40+ slightly complex functions. I just ordered and paid for one with three very small movements of one thumb.

One macro thing to keep an eye on in the distant future, like two years from now, is energy. Solar and, to a lesser extent, nuclear and wind are technologies. They are not sourcing/manufacturing/chemical process plays anchored on commodities. Their utility per unit dollar will follow the steep tech hockey stick curve. With AI doing what it is doing, the solar electric ecosystem will have 10x more value per unit spend in a few years. In under ten years, I would expect that from nuclear. Battery technology and advances in material science (AI-assisted) will be an additive part of that.

I will end with my favorite quote about technology and today from that Dickens guy:

It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity, it was the season of Light, it was the season of Darkness, it was the spring of hope, it was the winter of despair, we had everything before us, we had nothing before us…

#AI #GenAI #whitehouse #nationalcompetitiveness #innovation