How can you identify the best AI investment opportunities?

AI isn't just a Silicon Valley speed race, coded by tech bros. The real game-changer? Betting on companies steeped in industry smarts. Forget the dime-a-dozen algorithms—it's the seasoned industry veterans who train AI to be top-notch. That's where the gold is.

Look for businesses where the AI makes sense as a solution. Many companies claim to apply ML models to areas where either 1. The problem is not at a stage where it can be solved by AI or 2. The ML models are not yet advanced enough to solve the problem in the first place!

Why not start by visiting the computer department of universities close to you? It may take some time, but the next unicorn may be there.

D'après mon expérience, il faut que la Startup puisse vulgariser très simplement la solution qu'elle apporte et idéalement via un exemple très concret.

As the owner of a software product development company, I see clients at this crossroad almost every day. Yes, I understand, AI is the new hot thing to do, but no, not everyone needs it. You have to find the real pain point and challenge it to prove that AI is the answer. Most of the times that we were asked to implement AI, we ended up recommending an utomation workflow or RPA. As technology consultants, is our duty to advise our clients on the most efficient and cost effective way to address their requirements, and AI is ONE of many options.

In my quest to identify top AI investment opportunities, I've discovered that focusing on automation uncovers promising prospects. Delving into sectors primed for automation, like manufacturing or healthcare, yields valuable insights into areas where AI can significantly impact. Collaborating closely with industry insiders and staying attuned to emerging technologies have been instrumental in my investment journey. This personal touch ensures I can discern the most promising avenues for investment, aligning my decisions with the evolving landscape of automation and maximizing potential returns.

Investigate whether the startup has a technological edge in solving the problem. This could be proprietary data, a novel AI model, or a unique application of existing technology. Examine the diversity of use cases for the startup's solution. A16Z suggests that the best AI startups often have multiple potential applications, indicating a broad understanding of the problem landscape.

Adding 'AI' to your company's name is not enough. To pinpoint the most promising AI investment prospects, it's vital to evaluate the technology utilized or developed by the startup. Consider: What unique technical advancements does the startup bring? How effectively does it harness AI to address issues? What's the sophistication and reliability of its AI model or algorithm? How does it gather, handle, and apply data? How does it safeguard and capitalize on its intellectual assets? Seek out technologies that are innovative, viable, and secure, boasting a distinct competitive edge and value proposition.

Another key component is the team's ability to adapt to change. AI is rapidly changing at the moment- is the team able to pivot? Do they have the tenacity to completely turn the company around? Have they gone through what-if scenarios? What if someone copies them? What if big tech releases their exact tool?

If you have to choose between Product and Team, always prioritize the team. With a great team, you can develop any product. The most crucial aspect is having a well-structured, diverse team. Ideally, team members should possess diverse professional backgrounds and skills to complement each other. They should collaborate effectively, provide feedback, and tackle difficult situations together. One more thing - they should be passionate about what they are building and capable of making sound decisions promptly.

Assess the founders' passion for and expertise in the problem area. YC believes that founders deeply committed to solving a problem they've personally experienced are more likely to succeed. Their personal connection can drive innovation and persistence. The most effective teams combine deep technical knowledge with business acumen. Look for teams that have both: engineers and scientists who can develop and refine the AI technology, and business-oriented members who understand the market, can navigate the business landscape, and have sales and marketing expertise.

Especially with AI, team is the first thing that matters. You can use bespoke team evaluation and sourcing services like Olympos. The key is to meet the team early. Earlier than you think you should.

Team is top for me, always. Team has to execute idea and sell to customers. No compromises here and don't just get lost in whether the product idea is good.

Same principles in AI as any other business. Is there enough addressable market in the problem to make money. If you're solving a pain point for very small market segment, the profits are limited.

Everyone wants to invest in AI right now - very on-cycle in the parlance. But the market is changing really quickly and understanding the future direction is crucial to building a thesis. The land-grab in foundational models may now be over, except for very narrowly focused (maybe even small language) sector-specific tools. Now we're seeing the Cambrian explosion in services and applications that build on top of these foundational models. Are they all investible? Absolutely not. Wrapper companies will have 1000s of competitors. Be clear about what is unique technology, what leverages other platforms, and where the foundational players may want to build services as well.

Analyzing the market should also include assessing the narrative, there's a lot of hype around AI. Could that mean a plethora of entrants that are not yet visible but will soon compete with your investments? Will the firm genuinely build new AI capabilities or is it a fundraising gimmick?

Use the "Five Whys" technique to get to the root of the problem. Asking "why" repeatedly helps uncover the deeper market need or pain point the startup is addressing. A16Z advises investing in companies that solve a critical pain point, not a superficial one. Look for companies that have the potential to disrupt existing markets or create new ones. Assess if the problem the startup is solving could lead to market disruption. This often involves a unique approach or technology.

Stay abreast of emerging trends and developments in the AI industry. Monitor market reports, industry publications, and conferences to identify areas of rapid growth and innovation within the AI ecosystem.

Evaluate the clarity of the vision and how well it’s communicated. Leaders should be able to articulate where they see the company going and how they plan to get there. Previous successes, whether in starting a company or leading significant projects, can be a good indicator of future success. However, don’t discount the value of failure; many successful entrepreneurs have learned from past failures.

Assess the strength and uniqueness of the company's intellectual property portfolio, including patents, proprietary algorithms, and data assets. Strong IP protection can provide a competitive advantage and barrier to entry for competitors.

I talk to the founders. I try to read within the lines whether the startup has a product or absolutely nothing. After doing that for 12 years, it's easier to spot the difference. Start a conversation, and you'll know what I mean.

Assess the competitive landscape and identify potential competitors, partners, and disruptors in the market. Understand how the company's AI technology compares to competing solutions and what sets it apart in terms of performance, features, and value proposition.

➡️ Actively track academic research in core AI domains like machine learning, natural language processing, computer vision etc. Many startups spawn from cutting-edge papers. ➡️ Review data strategies and assess access to proprietary datasets that may serve as unfair competitive moats over time. ➡️ And the last but maybe the most important: join communities of leading enterprise AI adopters to connect with their innovators and understand pain points startups can solve.

In a field as rapidly changing as AI, continuous learning is essential. The team should be up-to-date with the latest developments in AI research and applications. The startup journey is rarely linear. The best teams are those that can pivot quickly in response to feedback, market changes, or new opportunities. Look for evidence of past pivots or strategic shifts in the company’s history. This indicates flexibility and the ability to adapt to changing circumstances.

Consider regulatory and ethical considerations related to AI technologies, such as data privacy, security, bias, and transparency. Look for companies that prioritize ethical AI practices and compliance with relevant regulations.