I’ve seen over a thousand pitches. Probably more. After my first exit with MovieLaLa, and especially after RemoteTeam was acquired by Gusto, my inbox became a firehose of ambition. Every founder comes armed with a deck, and every deck has that slide: the TAM, SAM, SOM breakdown. It’s the obligatory nod to market size, the chart that’s supposed to make investors like me feel comfortable that we’re not throwing money into a black hole.
Here’s my take: for AI companies, that slide is mostly garbage.
It’s a relic from a time when markets were static, well-defined, and waiting to be captured. You could calculate the Total Addressable Market (TAM) for shoes or enterprise software with some degree of confidence. But AI isn’t about capturing existing markets. It’s about creating entirely new ones. Trying to define the TAM for something that doesn’t exist yet is an exercise in fiction.
When I invested in Scale AI, what was the TAM for data labeling for autonomous vehicles? It was effectively zero. The market was a handful of research projects. Alex and his team didn’t enter a market; they willed one into existence through sheer force of will and a superior product. The same goes for my investments in Anthropic and OpenAI. What was the TAM for foundational models before they built them? It’s a nonsensical question.
The Great Illusion: Why Market Sizing Fails for AI
The traditional framework forces you into a box. It asks you to define your Serviceable Addressable Market (SAM) and your Serviceable Obtainable Market (SOM) as neat little subsets of a giant, imaginary number. This thinking is fundamentally flawed for AI startups for a few reasons.
First, AI creates new behaviors. Think about it. Before ChatGPT, how many people thought they needed an AI assistant to help them write emails or code? The market wasn't there. The technology itself created the demand. The most transformative AI companies don't just sell a better mousetrap; they invent the concept of mice and the need for trapping them. Your SOM isn't a percentage of the SAM; it's a beachhead from which you will redefine the entire TAM.
Second, technology commoditizes at a terrifying speed. I’ve seen it happen again and again. A team comes in with a brilliant new algorithm that gives them a 10% performance edge. Six months later, that edge is gone. A new paper is published, a new open-source model is released, and suddenly your core differentiator is a feature on someone else’s platform. If your only moat is a clever algorithm, you don’t have a moat. You have a puddle that’s about to evaporate.
This is the trap I see so many founders fall into. They are so focused on the technical brilliance of their solution that they forget to build a real, defensible business. They obsess over model performance and data pipelines while ignoring the one thing that can’t be easily copied: brand.
Brand is Your Only Real Moat
In the world of AI, the best technology doesn't always win. The brand that captures the market's imagination does. While everyone else is fighting in the trenches over a few percentage points of accuracy, the company with the strongest brand is rewriting the rules of the game.
When I talk about brand, I’m not talking about a slick logo or a catchy tagline. I’m talking about trust, reputation, and narrative. Why is it that developers flock to Hugging Face? It’s not just because they have a lot of models. It’s because they have built a brand around community, openness, and collaboration. They are the trusted hub for the entire ecosystem. That is a powerful position that can’t be replicated by simply spinning up a new model repository.
A strong brand in AI gives you three critical advantages:
Trust in the Black Box: Most AI models are effectively black boxes. Users and customers can’t peek inside to see how they work. They have to trust that the model will perform as advertised, that it won’t be biased, and that it will be secure. That trust isn’t given; it’s earned. A strong brand is a proxy for that trust. It’s the reason why enterprises will pay a premium for a model from a company like Anthropic, known for its focus on safety.
The Talent Magnet: The war for AI talent is brutal. The best researchers and engineers can work anywhere they want. Why would they choose your startup? Often, it’s the brand. They want to be associated with a company that is seen as a leader, that is pushing the boundaries, and that has a mission they believe in. My own experience as an angel investor in over 200 companies has shown me that the startups with the strongest brands consistently attract the best people.
Pricing Power and Community: When you have a strong brand, you’re not just selling a utility. You’re selling a vision. You’re selling access to a community. This allows you to command higher prices and build a loyal following that will stick with you even when a cheaper competitor comes along. Your users become your evangelists, your defenders, and your biggest source of feedback and innovation.
The Day One Playbook for Building a Brand
So how do you build a brand from day one? It’s not about a massive marketing budget. It’s about being intentional and authentic.
First, the founder is the first brand ambassador. You have to be the face of the company. You have to be the one out there telling the story, sharing your vision, and building relationships. I wrote Becoming Top 1% not just to sell books, but to codify my thinking and share it with a wider audience. It’s a branding tool.
Second, contribute, don’t just sell. The fastest way to build a brand in the AI space is to give back to the community. Open-source a tool. Publish your research. Write a blog post that actually teaches people something. When you contribute, you’re not just showing off your expertise; you’re building goodwill and demonstrating that you’re a part of the community, not just trying to extract value from it.
Third, have a strong, controversial opinion. The world doesn’t need another generic blog post about the potential of AI. It needs a point of view. Don’t be afraid to be provocative. My opinion on the uselessness of TAM slides for AI is a good example. It gets a reaction. It makes people think. And it positions me as someone who is willing to challenge conventional wisdom. What is your controversial opinion? What sacred cow of the AI world are you willing to challenge?
Fourth, build in public. Share your journey. Talk about your successes and your failures. Document the process of building your company. This creates a narrative that people can follow and connect with. It makes your company more human and relatable. People don’t just buy products; they buy into stories. Make your story a compelling one.
The Bottom Line
Stop wasting your time on meaningless market sizing exercises. The market for truly innovative AI is unknowable because you are the one who is creating it. Your job is not to fit into a pre-existing box but to build a new one.
Instead of obsessing over TAM, SAM, and SOM, obsess over your brand. Your brand is your story. It’s your reputation. It’s the reason why the best people will want to work with you and why customers will trust you with their business. In an industry where technology is constantly in flux, your brand is the one thing that can provide a lasting, defensible advantage.
So, the next time you’re putting together a pitch deck, I challenge you to do something different. Take that TAM, SAM, SOM slide and throw it in the trash. Replace it with a slide that tells me about your brand. Tell me about the story you’re telling, the community you’re building, and the dent you’re going to make in the universe. That’s a story I’ll invest in.
Frequently Asked Questions
What experience informs this perspective?
This perspective comes from over a decade of building companies in Silicon Valley, two successful exits (RemoteTeam to Gusto, MovieLaLa to Gfycat), and investing in 200+ startups including Anthropic, OpenAI, and Scale AI. I write about what I've lived.
How has this view evolved over time?
My thinking on most topics has changed significantly over the years. Early in my career, I held many conventional views that experience proved wrong. I try to update my beliefs when the evidence changes.
Do all experts agree with this view?
No, and that's fine. The best ideas in business are often contrarian. I share my perspective based on my experience and data, but I encourage you to seek out opposing viewpoints and form your own conclusions.
What's the most common pushback you get on this?
People often push back by citing exceptions or edge cases. And they're usually right that exceptions exist. But building a strategy around exceptions rather than patterns is a losing game for most founders.