I’ve seen more AI pitch decks than I can count. Hundreds, maybe thousands by now. And I’ll tell you a secret: most of them are garbage.
That might sound harsh, but it’s the truth. We’re in the middle of a gold rush, and everyone is slapping an "AI" label on their company hoping to strike it rich. I’ve been an entrepreneur, and I’ve had two successful exits. I’ve also been an angel investor in over 200 companies, including some of the biggest names in AI like Anthropic, OpenAI, Scale AI, and Hugging Face. I’ve seen what it takes to build a real, lasting AI company. And I’ve seen the warning signs that tell me a startup is destined to fail.
So, I’m going to share my personal checklist of red flags. These are the things that make me close a pitch deck and move on to the next one. If you’re a founder, this is your guide on what to avoid. If you’re an investor, this is your guide on what to look for.
Red Flag 1: “AI” as a Buzzword, Not a Business
The first thing I look for is whether a company is a true AI company or just a company using AI. There’s a huge difference.
A true AI company has a core, proprietary technology that gives them a competitive advantage. They’re not just using a third-party API. They’re building something new, something that’s hard to replicate.
I remember one startup that pitched me a few years ago. They claimed to have a revolutionary AI-powered platform for e-commerce. It sounded impressive. But when I dug into their tech, I found out they were just a thin wrapper around a popular recommendation engine’s API. They had no secret sauce, no defensible moat. They were a marketing company, not a tech company. I passed.
It’s not that using third-party APIs is bad. It can be a great way to get started. But if your entire business is built on someone else’s technology, you’re not a true AI company. You’re a reseller. And that’s a tough business to be in.
Red Flag 2: The Data Desert
AI is nothing without data. You can have the most brilliant algorithm in the world, but if you don’t have the data to train it on, you have nothing. That’s why the second red flag I look for is a lack of a clear data strategy.
I want to see that a company has a plan for acquiring a unique and valuable dataset. This could be through partnerships, user-generated content, or a clever data acquisition loop built into their product. What I don’t want to see is a company that’s planning to just scrape the web or use publicly available datasets. That’s not a sustainable advantage.
I once met a team with a fantastic idea for a personalized medicine platform. Their algorithm was top-notch. But they had no data. They were hoping to partner with hospitals to get access to patient records, but they had no agreements in place. They were stuck in a chicken-and-egg problem: they couldn’t get the data without a product, and they couldn’t build a product without the data. I wished them luck, but I didn’t invest.
Red Flag 3: The “Business Guys” Who Can’t Code
AI is a deeply technical field. You can’t fake it. That’s why I’m always wary of teams that are heavy on business and marketing experience but light on technical talent.
I want to see a team of founders who have a deep understanding of the technology they’re building. At least one of the founders should be a world-class engineer or researcher. If the founders are all “idea people” who are planning to outsource the development, that’s a huge red flag for me.
I’m not saying that business and marketing aren’t important. They are. But in an AI startup, the technology has to come first. I once listened to a pitch from a team of two co-founders. They were both incredibly charismatic and had impressive backgrounds in sales. But when I started asking them technical questions about their product, they couldn’t answer them. They kept deferring to their “tech team,” which consisted of a few freelance developers they had hired on Upwork. I passed. I need to see the passion and the expertise in the founding team.
Red Flag 4: TAM-tastic Delusions
Total Addressable Market (TAM) slides are a running joke in the VC community. Every founder thinks their TAM is in the billions, or even trillions. But most of the time, these numbers are pure fantasy.
I want to see a realistic, bottom-up analysis of the market size. I want to see that the founders have a clear understanding of who their customers are and how much they’re willing to pay. I don’t want to see a top-down analysis that starts with a huge market and then claims to be able to capture a small percentage of it.
I’ll never forget one pitch deck that claimed their TAM was the entire global population. They were building a consumer app, and they argued that everyone on Earth was a potential user. That’s not a TAM, that’s a dream. A real TAM is based on a specific customer segment with a specific need that your product can solve.
Red Flag 5: The Missing Moat
In the world of investing, a “moat” is a sustainable competitive advantage that protects a company from its rivals. In the world of AI, moats are more important than ever. The technology is moving so fast that any new innovation can be quickly copied.
That’s why I’m always looking for companies that have a clear path to building a defensible moat. This could be through network effects, a proprietary dataset, a strong brand, or a unique technology. What I don’t want to see is a company that’s competing on features alone. That’s a race to the bottom.
I invested in a company a few years ago that had a great product but no moat. They were quickly overtaken by a larger competitor who copied their features and out-marketed them. It was a painful lesson, but it taught me the importance of looking for a sustainable competitive advantage from day one.
The Bottom Line
Building a successful AI startup is hard. It’s not enough to have a great idea or a great team. You need to have a real, defensible business. You need to have a clear vision for how you’re going to win.
So, if you’re a founder, I hope this list of red flags has been helpful. And if you’re an investor, I hope it’s given you a few things to think about. The AI revolution is just getting started, and there are going to be some massive winners. But there are also going to be a lot of losers. My job is to tell the difference between the two. And now, hopefully, you can too.
Frequently Asked Questions
What are the most common mistakes when spoting warning signs in ai startups as a vc?
The biggest mistake I see is overcomplicating things early on. Start with the simplest version that works, get real feedback, and iterate from there. Another common trap is copying what worked for someone else without understanding the context behind their decisions.
What tools do I need to get started?
Start with the basics. You don't need expensive software or fancy tools. A spreadsheet, a note-taking app, and direct access to your customers will get you further than any enterprise platform. Add tools only when you hit a specific bottleneck.
How long does it take to spot warning signs in ai startups as a vc?
The timeline varies depending on your starting point and resources. For most founders, expect 2-4 weeks for initial setup and 2-3 months to see meaningful results. I've seen teams move faster when they focus on one thing at a time rather than trying to do everything at once.
How do I measure success with this approach?
Pick one or two metrics that directly tie to your goal and track them weekly. Vanity metrics like page views or follower counts rarely matter. Focus on metrics that reflect real engagement or revenue impact.