An AI summer

Israel-based iAngels is one of the world’s most significant venture capital and investment platforms. With $400 million AUM and 110 companies in its portfolio, it provides investors access to Israeli innovation. Here, Shelly Hod Moyal, founding partner of iAngels, talks to CampdenFB about where she sees the opportunities in artificial intelligence (AI), the need for government regulation and ponders advice for AI entrepreneurs. 

AI is one of the most exciting fields for venture capitalists at the moment. What types of AI are exciting you at the moment?

Shelly Hod Moyal: At iAngels we’re excited about AI generally and the impact it has across so many industries, however it’s important to filter through the hype as many founders use the word “AI” just to add flair. 

While every company could use AI in some part of their value proposition, we’re looking for teams that can disrupt markets in a profound way while maintaining a defensible barrier. This might be found in the following sectors:

• Exclusive access to proprietary data that fuels AI models creating a network effect, where the AI improves as more data is added. This makes it harder to replicate, even with access to more advanced AI models.
• Deep expertise in specific sectors like cybersecurity, fraud detection, healthcare diagnostics, and enterprise applications which can give a unique competitive edge. Specialised AI models that focus on these areas are likely to outperform more generalist models, making it difficult for competitors to keep up.
• AI infrastructure, such as DevOps tools, MLOps, and cloud infrastructure, that would serve underlying layers required to support the growth of the AI as a whole.
• AI applications which could democratise and automate business processes in ways that transform SMBs and consumer markets, leading to a real paradigm shift in how businesses operate.

What particular fields is iAngels looking at? 

Shelly Hod Moyal: We’re primarily looking at the following AI related sectors: AI in cyber security, AI in enterprise software (SaaS) and DevOps/Cloud Infrastructure, and AI in healthcare. 

One particularly interesting company is Lasso Security, which is positioning itself to be a leader in AI-driven cybersecurity. Lasso focuses on protecting large language models (LLMs) and generative AI systems from advanced threats like prompt injection attacks and model theft. As businesses increasingly adopt AI, safeguarding these technologies becomes crucial, and Lasso’s solution provides end-to-end protection for companies leveraging AI.

Another fascinating company is Immunai, which is using AI to personalise treatments, particularly immunotherapy for cancer patients. This kind of tailored medicine represents a major leap forward. Although the primary focus has been cancer, this approach can be expanded to other conditions as well. They actually just partnered with AstroZeneca in an $18m deal to drive efficiency in drug development through Immunai’s AI model. 

The potential for personalised medicine here is groundbreaking. What seemed unimaginable just a decade ago is now not only possible but already in practice. It’s now a question of scaling these innovations and bringing them into wider production.

In addition, we have startups working more on the software side of healthcare. One promising company is Binah.ai, which uses AI to monitor vital signs with just a smartphone. Unlike wearables like the Apple Watch, which track metrics like blood pressure and heart rate, Binah.ai can also measure blood oxygen saturation, and they’re currently developing the ability to track glucose levels – all remotely, without the need for physical contact. 

This technology can revolutionise the future of healthcare, allowing for remote diagnostics and triaging using just a camera. The implications for accessibility and efficiency in medicine are extraordinary, and it’s all made possible through AI.

How do you avoid rabbit holes? How do you stop yourself getting so excited by the technology for something that has no commercial application? 

Shelly Hod Moyal: For us, the focus is always on identifying market opportunities. In the past, coding and technical skills, such as computer engineering, gave startups a competitive edge. However, with the advancements in AI, coding has become faster and more accessible. As a result, many aspects of technology – especially in terms of architecture and development – have become or are in the process of becoming commoditised. This shift makes business model innovation, productisation, and go-to-market strategies far more critical to a startup’s success.

If anyone can now code quickly and efficiently, the challenge isn’t just in developing the technology – it’s in how you position it. Once others catch on to what you’re doing, they can replicate it just as fast.

That’s why our focus as investors is always on the market: how the technology can be used, how defensible the solution is, and how “sticky” it becomes for users. As an investor, I’m always considering the market – looking beyond the technology itself to understand how it can truly create long-term value. We seek companies that have demonstrated early traction, with teams that are intensely focused on engaging design partners early in the market research and product development process in order to accelerate the path to product-market fit.

What is it that entrepreneurs are typically missing beyond money? 

Shelly Hod Moyal: There’s no one-size-fits-all approach when it comes to entrepreneurship. Every entrepreneur has their own unique path and challenges. Some may have already achieved exits in a different sector, while others might need help building a team or shaping a business model. Building a startup is vastly different from being an employee. As an employee, you can put your head down, work hard, and advance, but as an entrepreneur, you can’t go it alone. It’s not enough to have a solid team of developers – you also need capital, partners, talented employees, and, ultimately, everything from customer satisfaction to market timing has to align for success. Humility is also a valuable trait that helps entrepreneurs address blind spots in management, market perception, investor relations, and other key areas.

For early-stage entrepreneurs, investor support is crucial. Many founders need guidance on setting up the foundations of corporate governance, negotiating with strategic partners, recruiting talent, and managing teams. Every entrepreneur can benefit from coaching, but the role of the investor is especially important as a sounding board. Some startups need more help with networking and connections, while others require a different type of support.

Ultimately, there’s no universal formula for success. Each entrepreneur’s journey is unique, and the type of support they require varies from one to the next.

Do you have the sense that governments understand AI? 

Shelly Hod Moyal: A lot of people, including governments, entrepreneurs, and even technologists, don’t fully understand AI. There’s this general belief that we can build and control it, but as AI becomes increasingly intelligent, it’s important to be more thoughtful. There’s a real possibility that AI could surpass human intelligence, leading to second, third, and fourth-order consequences that no one has anticipated. 

Any decision you make can have a “butterfly effect”, where small actions can result in significant, unforeseen consequences. Overall, I don’t think anyone truly understands where AI is taking us, or what the world will look like several years from now. The complexity and unpredictability of AI’s evolution mean that we’re navigating largely uncharted territory. 

It is critical for entrepreneurs to actively work to make new AI capabilities safe. One unique example is Ilya Sutskever, former Chief Scientist at OpenAI, who left his secure position to build Safe Superintelligence or SSI, an LLM whose name and mission is safety first. 

Are you finding regulation helpful with AI, and does that have any impact on how you invest?

Shelly Hod Moyal: Right now no, there is very limited regulation. I don’t think people fully appreciate the potential dangers of AI yet. Similar to the early days of the atom bomb – humanity didn’t truly grasp its destructive power until we saw its effects firsthand. Once we understood the gravity of it, there was an urgent need for coordination, treaties, and regulations. Even then, while we managed to create some frameworks around nuclear weapons, we’ve seen the proliferation of bombs and their increasing danger, with more governments and agencies gaining access to the technology.

With AI, the situation is even more complex and alarming. Governments, by nature, are rarely aligned or coordinated, making it difficult to control the advancement and use of such technology. AI in a sense is particularly challenging to contain because it is software based and largely open source with technology that is accessible to the broad public. Given how incredibly powerful AI is becoming, I believe we will need government regulation and international coordination. Institutions will need to be established to set best practices and monitor them. Additionally, there will need to be a significant level of corporate responsibility from the major players in AI development – companies like Nvidia, Microsoft, Google, Meta, and OpenAI, as well as those building large language models (LLMs). These corporations will play a crucial role in ensuring that AI is used safely, and that we avoid or at least mitigate the potential for catastrophic outcomes.  

How do you personally manage those ethical concerns? 

Shelly Hod Moyal: As with every technology, the more powerful the technology, the greater the concern for misuse. The first step in managing concerns happens in the DD process when we analyse an opportunity and the risks associated. Assuming we believe the potential reward outweighs the risks, we try to determine how to mitigate those risks.

A key mitigating factor can be the team, if its members are strong, ethical, experienced and responsible. We spend a lot of time in our DD trying to determine the strength of the team, conducting thorough background checks, psychological assessments, personality tests and reference checks for each founder. We consider how they treat their team, partners, how they manage, research and negotiate – analysing their leadership style and ability to adjust and adapt in an uncertain environment. We believe that to adapt, entrepreneurs need to possess humility so they will take necessary precautions and hire professional advice in areas they don’t know.

Another mitigating risk is the investor’s involvement with the company, where they are able to provide guidance on what the entrepreneurs may not see – whether that’s preparing for regulatory hurdles by securing legal counsel or safeguarding against risks, both technological and regulatory. This oversight is crucial to helping startups navigate the complexities of growth as well as avoiding pitfalls.

What advice would you give to an entrepreneur coming to you with an AI startup? 

Shelly Hod Moyal: Given the rapid commoditisation of technology and AI’s ability to expedite coding, it’s crucial for entrepreneurs to focus sharply on the product itself. They need to think strategically about how to bring that product to market. It’s not just about being the first to market but also about making the product “sticky” so that customers remain loyal, especially since competitors can quickly copy what works well.

Another essential factor is ensuring their product is language-agnostic, meaning it should not be built solely on one language model, like ChatGPT. Entrepreneurs should design infrastructure that can work across different language models because there’s uncertainty about which model will dominate in the future. By diversifying their approach, they avoid limiting themselves to a specific model or platform, and position themselves to adapt to future shifts in AI technology.

Finally, they should look for opportunities in areas where the largest tech companies – especially the so-called “Magnificent Seven” – aren’t focusing their efforts. By identifying niches that these giants are less likely to enter, startups can either build sustainable businesses or position themselves for acquisition. Understanding the strategic plans of major players can help entrepreneurs innovate in areas where they can either avoid direct competition or offer something that would make them attractive for a potential buyout. 

These three elements – strategic product focus, language model flexibility, and niche innovation – are key to thriving in an AI-driven market.

The image at the top of the story is of Shelly Hod Moyal and Mor Assia. All photos copyright Efi Sameach