M&A activity in the AI sector is booming and Israel is a major player. From autonomous systems to predictive cybersecurity, Israeli-founded AI startups are at the forefront of global innovation. As both strategic acquirers and financial investors race to secure emerging technologies, a distinct category of transactions is gaining ground: cyber-AI deals that combine machine learning, threat detection, and large-scale data processing.
But acquiring an AI company presents a very different set of challenges compared to traditional software or tech businesses. These deals require a deeper understanding of the target’s technology, risk profile, and market dynamics – particularly when it comes to the following key areas:
1. Algorithm ownership Is often ambiguous
AI models are rarely developed in isolation. They often draw on a mix of proprietary code, open-source tools, customer data, and even academic research. This creates real complexity around who owns the resulting model-and whether the company has the rights to commercialize and license it. Disputes over IP ownership, licensing limitations, or unclear contributions can undermine the value of a deal if not thoroughly vetted.
2. Regulatory uncertainty adds risk
AI regulation is evolving rapidly with the EU AI Act, US policy proposals, and new frameworks emerging globally. Many AI companies, especially early-stage startups, lack formal compliance structures. Buyers need to assess not only current compliance, but also the company’s ability to adapt to future legal and ethical standards. This mismatch between today’s diligence and tomorrow’s liability makes regulatory risk a growing concern in AI transactions.
3. Revenue concentration is a common theme Unlike more mature SaaS companies, AI startups often depend heavily on a handful of customers or pilot projects. That means any misrepresentation about the status, renewability, or terms of key contracts can have an outsized impact. Understanding the durability of those relationships- and their legal enforceability – is essential.
4. Cyber risks can be hidden even in cyber companies
Ironically, companies specializing in cyber-AI may themselves be exposed to cybersecurity issues. Past breaches, poor data governance, or overstated claims about compliance with security standards can lead to unexpected liabilities. It’s critical to go beyond surface-level checks and assess whether security practices align with customer expectations and legal obligations.
Looking ahead
With Israel’s AI and cybersecurity sectors increasingly overlapping, the surge in cyber-AI M&A is expected to accelerate. These deals offer exciting opportunities – but also demand careful navigation of risks that are still poorly understood in many corners of the market.
As these complexities grow, dealmakers are responding with smarter diligence and tailored protections. One tool increasingly used in this space is Representations & Warranties Insurance (RWI), which can offer recourse for unknown breaches of key contractual representations, including those tied to IP, regulation, and cyber.
Ultimately, successful AI M&A requires more than just traditional deal-making. It calls for a cross-disciplinary approach that blends legal, technical, and commercial insight. For those who get it right, the upside is significant.
The author is a Head of M&A at WTW (Willis) Israel. He specializes in advising global buyers and Israeli sellers on the use of M&A insurance in domestic and cross-border transactions.
Published by Globes, Israel business news – en.globes.co.il – on July 28, 2025.
© Copyright of Globes Publisher Itonut (1983) Ltd., 2025.
