AI is not one race, but dozens of races happening simultaneously across three domains: Infrastructure, Intelligence, and Interface. Each race has its own competitors, its own investors placing bets, and its own commentators narrating from the sidelines. The reason most people feel overwhelmed is not that AI is too complex to understand. It is that they are hearing the noise from all three layers at once, with no framework for sorting it. This piece provides that framework by identifying the three domains, the players inside each one, and then giving you three simple questions that will filter out most of the noise before you ever react to a headline.

---

For the record, I am optimistic about AI. I believe it has and will unlock real productivity gains, and reshape the future of work. I believe it will change the structure of organizations over the next decade in ways most people still underestimate. And I believe the finish line of this entire movement is the self-managing organization: a company that can operate, learn, and adjust with minimal human supervision, no different in concept from the self-driving car, but infinitely more complicated in practice.

I also believe the hype is beyond excessive. The volume of commentary surrounding AI right now is extraordinary. Founders are promising revolutions. Investors are forecasting trillion dollar outcomes. Consultants are declaring existential urgency. Influencers are distilling complex developments into viral narratives. Vendors are repositioning existing products as AI native. Media cycles amplify every breakthrough and every setback in alternating waves of euphoria and panic.

When I walk into a very noisy room, I stop talking and I watch. I look at who is speaking and who others are listening to. I notice who gathers around whom. I pay attention to who nods, who rolls their eyes, who interrupts, who defers. I watch who benefits from the direction the conversation is moving and who grows uncomfortable. I map incentives, observe alliances, and identify positioning. And every time, once I understand who is who and what they want, I can see exactly what is happening.

Right now, AI is happening in a room where everyone is talking at once. People are pitching, selling, fundraising, warning, worrying, promoting, celebrating, often with equal conviction and completely different incentives. If you follow the loudest voice, you will mistake volume for truth. Not because AI is fake, but because the room has not been mapped yet: who is speaking, who benefits, and what game is actually being played.

My goal with this article is to help you map the room. I intend to explain what is actually happening, identify the players involved, and give you a simple structure for interpreting events as they unfold. Not so you can predict the future, but so you can stop feeling like everyone else understands something you do not. Because once you can see the field, you do not need a PhD in machine learning to make sense of it. Your common sense will be sufficient.

Full disclosure: I am not a neutral observer. My own companies are fully committed to building toward that finish line. I have spent years developing the theory, the software, and the services required to move organizations toward autonomy. I am a player in this race, and I will be transparent about that throughout everything I publish. I tell you this now so you can weigh my perspective accordingly, which is exactly the skill I am about to teach you.

---

Takeaways:

1. AI is real and transformative, but the hype surrounding it is distorting the signal.
2. The finish line of this movement is the self-managing organization, and the race to get there will truly reshape every industry.
3. Most of the confusion people feel comes from consuming AI discourse without understanding who is speaking and what they want.
4. The goal of this article is to give you a simple, reusable framework for cutting through that noise.
5. The author is a participant in this race, not a neutral observer, and will be transparent about that throughout.

It Is a Race. Treat It Like One.

When I was a kid, we would go to the sports complex on weekends to play baseball. There were many fields, each hosting a different game with its own teams, its own score, its own crowd of parents and spectators in the bleachers, sometimes a guy in the press box calling the play-by-play, and always a few people on the side making bets on who would win. You could stand in the parking lot and hear the noise from all of them at once, and it would sound like chaos. But walk up to any single field and the game made perfect sense. Two teams. A set of rules. A score. A winner.

Right now, AI is just like that, except there are three sports complexes, not one. Each complex has its own set of fields. Each field has its own game in progress with its own competitors racing for the trophy. Every game has its own backers in the bleachers rooting for the teams they have invested in, and its own commentators narrating the action from the sidelines. And if you stand in the middle of all three complexes and try to listen to everything at once, it sounds like total chaos. That is the experience most people are having right now.

The reason it feels overwhelming is not that AI is too complex to understand, but that you are hearing the noise from dozens of games across three different complexes, all at the same time, with no map telling you which complex you are listening to or which game matters to you. I call these three complexes Infrastructure, Intelligence, and Interface. Every company, product, investment, influencer, and headline you encounter in AI belongs to one or more of them. Once you know which complex you are looking at, all this noise starts to sort itself.

The three complexes are:

1. Complex 1: Infrastructure. This is the physical and computational foundation of everything happening in AI. Hardware. Chips, data centers, networking, energy, cloud capacity. The races here are measured in silicon, watts, and capital expenditure. Think NVIDIA, AMD, TSMC, AWS, Microsoft Azure, Google Cloud. The racers in this complex are building the ground the rest of AI runs on.
2. Complex 2: Intelligence. This is where models are trained, benchmarks are chased, and reasoning ability is developed. The foundation model companies, the research labs, the open-source collectives. This is the complex that generates the most headlines and absorbs the most public attention. It is also the complex most likely to commoditize over time. Think OpenAI, Anthropic, Google DeepMind, Meta AI, Mistral, DeepSeek.
3. Complex 3: Interface. This is where intelligence meets the real world. Where software actually changes how people work, how organizations operate, and how value is delivered. This complex will ultimately determine whether AI transforms your business or remains an expensive experiment. Think Salesforce, ServiceNow, Microsoft Copilot, and a rapidly forming field of new entrants, including my own companies Prospus and Kaamfu. The racers here are the most numerous and the hardest to see clearly because the field is still forming.

Each of these complexes has dozens of games in progress. Within Infrastructure alone, there are races for GPU dominance, chip architecture, data center capacity, and energy supply. Within Intelligence, there are races for model performance, reasoning capability, cost efficiency, and open-source parity. Within Interface, there are races for the platforms people will use every day, the tools that will change how work gets done, and the products that turn AI from a concept into something you can actually use.

These are not metaphorical races. They are real competitions with real winners and real losers, playing out right now across all three complexes simultaneously. Billions of dollars are being wagered and entire corporate strategies are being restructured around the outcomes. Careers, companies, and in some cases national economic agendas are riding on which racers pull ahead and which fall behind. And the vast majority of people watching have no framework for telling the difference between a real lead and a good press release. That is what we are going to fix.

---

Takeaways:

1. AI is not one race. It is many races happening simultaneously across three complexes: Infrastructure, Intelligence, and Interface.
2. Each complex has its own racers competing for dominance, its own backers and bettors wagering capital on outcomes, and its own commentators narrating the action.
3. The reason AI feels overwhelming is that most people are hearing the noise from all three complexes at once, with no map.
4. Infrastructure is the physical foundation: chips, data centers, energy, cloud. Intelligence is the capability layer: models, benchmarks, reasoning. Interface is where it all meets the real world: the software that actually changes how work gets done.
5. Once you know which complex a headline belongs to and who is speaking within it, the noise becomes manageable.

The Races

Now that you can see the three complexes, the next step is understanding who is inside each one and what role they play. Every race, in every complex, has three groups of people worth paying attention to. Learn to identify them and you will never be confused by an AI headline again.

These three groups are the Racers, the Backers, and the Commentators. They exist in every race within a complex, and each group has fundamentally different incentives. Confusing one for another is where most people go wrong. A commentator’s prediction is not a racer’s roadmap. A backer’s enthusiasm is not proof of a racer’s viability. And a racer’s press release is not the same thing as a racer’s actual product or performance. Once you can sort who is who, which races they are in and which complexes they belong to, the entire landscape becomes easier to navigate.

What follows is a breakdown of all three groups across all three complexes. This is not meant to be exhaustive. The players are changing constantly, and new entrants appear every week. The purpose here is to show you the structure so that when a new name appears or a new deal is announced, you already know where to place it and how to evaluate it. The names will change, but the structure will not.

---

The Racers.

These are the companies and teams actually racing. They are on the track, competing. They have products, they have roadmaps, they have payroll, and they have something real at stake beyond commentary.

• In Infrastructure, the racers include the chipmakers like NVIDIA, AMD, Intel, and emerging challengers like Cerebras and Groq fighting for a piece of the GPU race. On the cloud and data center side, AWS, Microsoft Azure, and Google Cloud are racing for compute dominance while companies like CoreWeave and Lambda Labs are carving out positions underneath them. TSMC sits in a unique position as the manufacturer that most of the chip racers depend on. And behind all of it, there is an energy race: the companies building the power infrastructure that AI’s appetite for electricity demands, from nuclear startups to utility-scale solar providers cutting deals directly with data center operators.
• In Intelligence, the racers are the foundation model companies: OpenAI, Anthropic, Google DeepMind, Meta AI, and Mistral at the front of the pack. But the open-source movement is a racer too, and an increasingly dangerous one for the incumbents. DeepSeek, LLaMA, and a growing community of open-weight model developers are closing the performance gap faster than most people expected. This complex also includes the companies building specialized models for specific domains: medical reasoning, legal analysis, code generation, scientific research. Not every Intelligence racer is trying to build the biggest general model. Some are racing to own a vertical.
• In Interface, the field is the widest and the least settled. The established enterprise players like Salesforce, ServiceNow, and Microsoft with Copilot are racing to embed intelligence into existing workflows. Vertical AI startups are racing to own specific industries. And a new generation of companies, including my own, Prospus and Kaamfu, are racing to build what I believe will become the autonomous operating environments of the future: unified platforms where intelligence, structure, and execution converge into a single working environment. This is the complex where the most value will ultimately be created, and it is also the one where the field is still widest open.

Not every racer is running the same race, and not every racer is running with the same strategy. Some are sprinting for dominance. Some are running a longer race toward durability. Some are running in the wrong lane entirely and will not realize it for another year. Knowing who is running where, and why, is the first step toward interpreting anything you read about AI.

The Backers and Bettors.

These are the investors, the corporate partners, the sovereign funds, and the strategic allies putting capital behind specific racers. They are not on the track themselves, but they have enormous influence over who gets to keep running and for how long. Every major AI company has backers, and those backers have their own incentives that do not always align with the racer’s stated mission.

• In Infrastructure, the bets are staggering. Sovereign wealth funds in the Middle East and Asia are funding national chip fabrication capacity. Private equity firms are pouring capital into energy infrastructure. The hyperscalers, AWS, Azure, and Google Cloud, are spending tens of billions per quarter, which makes them both racers and bettors simultaneously, funding their own position on the track.
• In Intelligence, the backing relationships are some of the most consequential in the history of technology. Microsoft has invested over thirteen billion dollars in OpenAI. Amazon has committed billions to Anthropic. Google funds its own DeepMind internally. Venture capital firms like Sequoia, Andreessen Horowitz, and Thrive Capital are placing massive bets on which model company will own the intelligence layer. These are not passive investments. They are strategic bets that shape product direction, pricing, partnerships, and the competitive dynamics of the entire complex.
• In Interface, the bets are more distributed but no less strategic. Enterprise-focused venture firms are funding the next wave of workplace AI. The strategic investment arms of companies like Google, Salesforce, and SAP are placing bets on which interface will own the workflow of the future. And increasingly, the Intelligence racers themselves are becoming backers in the Interface complex, offering credits, partnerships, and integrations to the application companies that will distribute their models into the real world.

When you see a headline about a massive funding round or a new partnership, do not just read the announcement. Ask who is backing whom, and what that backer gains if the racer wins. The backing relationships reveal more about where AI is actually going than any product demo.

The Commentators.

These are the analysts, the consultants, the conference speakers, the media personalities, and the social media voices who narrate the race from the sidelines. They are not racing. They are not funding. But they shape perception in powerful ways, and they have their own incentives: attention, consulting revenue, event tickets, thought leadership positioning, and media placement.

• In Infrastructure, the commentators include semiconductor analysts at firms like Gartner and IDC, financial analysts covering NVIDIA’s every earnings call, and an entire ecosystem of tech journalists tracking the GPU supply chain as if it were a geopolitical event.
• In Intelligence, the commentators are the loudest of all: AI researchers with large social media followings, benchmark leaderboard trackers, conference organizers at NeurIPS and TED, and a daily cycle of media coverage that treats every model release as a civilizational turning point.
• In Interface, the commentator class is dominated by the consulting firms, McKinsey, Deloitte, Accenture, and BCG chief among them, along with enterprise analysts, productivity influencers, and the entire conference circuit selling transformation frameworks to anxious executives.

This is not a criticism. Commentary is valuable when it is grounded. But when you are trying to make sense of AI, it matters whether the voice you are listening to has skin in the game or is selling tickets to watch. Most of the noise in AI right now comes from the commentator class, and most of it is calibrated to generate engagement rather than clarity. The commentator who tells you the world is about to end and the commentator who tells you everything will be fine are often driven by the same incentive: your attention.

---

That is the landscape. Three complexes. Dozens of races. And in every race, three groups of participants with fundamentally different roles and fundamentally different motivations. The racers are building. The backers are wagering. The commentators are narrating. None of them are lying to you, necessarily, but all of them are speaking from a position, and that position shapes everything they say.

The point of this framework is not to make you cynical. It is to make you literate. When you can look at any AI headline and immediately place it in a complex, identify whether the source is a racer, a backer, or a commentator, and understand what incentive is driving the message, you are no longer consuming the news. You are reading it. And that difference will matter more than almost anything else as you figure out what AI means for your own organization.

---

Takeaways:

1. Every race in AI has three groups: Racers (the companies actually competing), Backers and Bettors (the capital behind them), and Commentators (the voices narrating from the sidelines).
2. Each group has different incentives. Do not confuse a commentator’s prediction for a racer’s roadmap, or a backer’s enthusiasm for proof of viability.
3. In Infrastructure, the races span chips, cloud, data centers, and energy. In Intelligence, they span foundation models, open-source, and specialized verticals. In Interface, they span enterprise platforms, vertical AI, and the emerging autonomous operating environments.
4. Some of the biggest players are both racers and bettors simultaneously, funding their own position on the track. These dual roles reveal where the real conviction is.
5. Who is funding whom and why (the backing relationships) tell you more about where AI is heading than any product announcement or media headline.
6. Most of the noise in AI comes from the commentator class. Commentary is valuable when grounded, but always ask whether the voice has skin in the game or is selling tickets to watch.

Why This Matters for You

You are in a race too. Whether your organization has acknowledged it or not, you are in a race to restructure operations, integrate intelligence into the way work actually happens, and build the kind of organizational architecture that will define competitive advantage for the next decade. That race is yours to run. But to run it effectively, you need to understand the larger race you are running inside of. A misunderstanding at the beginning will do the most damage.

If you mistake an Infrastructure story for an Intelligence story, you will draw the wrong conclusions. If you take a commentator’s forecast as a racer’s roadmap, you will misallocate attention and resources. If you do not understand where the capital is flowing and why, you will be surprised by outcomes that were visible to anyone who was watching the backers. And if you react to every headline without knowing which complex it belongs to or who is speaking, you will exhaust yourself chasing noise.

The exercise is simple. From this point forward, when you encounter any AI news, any announcement, any prediction, any opinion, ask three questions before you react.

1. Which complex is this about? Infrastructure, Intelligence, or Interface?
2. Who is speaking? A racer, a backer, or a commentator?
3. What is their incentive? What outcome benefits them? What are they selling, funding, or promoting?

If you can answer those three questions, you have already filtered out most of the noise and you are no longer a passive consumer of AI discourse. You are an informed observer with a map.

In the posts that follow, I will walk through each complex in detail. I will go deeper into the racers. I will trace the backing relationships. I will explain the economics of each race and where the leverage actually sits. And I will provide ongoing commentary as events unfold, always returning to fundamentals before forming conclusions.

AI is powerful. But it is also a market, a capital cycle, a narrative engine, and a strategic battlefield. If you can see all of those dimensions at once, the story becomes far more revealing and far less confusing. Your journey does not need to be a carnival. It can be informed, intentional, and deliberate. And it starts right here, with understanding the race.

---

Takeaways:

1. You are already in this race. The question is whether you are running it informed or running it blind.
2. Before reacting to any AI news, ask three questions: which complex, who is speaking, and what is their incentive.
3. Those three questions alone will filter out the majority of the noise and protect you from misallocating attention, resources, and strategy.
4. In the posts ahead, each complex will be explored in depth with named racers, mapped backing relationships, and an honest read on where the leverage sits.