As artificial intelligence (AI) continues to revolutionize the business landscape, its associated risks are becoming more complex, widespread, and consequential. While the insurance industry determines the precise circumstances in which insurance may cover these risks, businesses should consider the complementary benefits of indemnification agreements as gap fillers.

The Rising Tide of AI-Related Risks

As we have explained before, AI risks are many and continuously evolving, posing significant and increasing challenges for businesses. As AI systems become more integrated into every part of the economy, businesses will face increased exposure to AI-enhanced liabilities running the gamut from cybersecurity breaches, data privacy violations, and product liability claims to board and management-level liabilities based on misstatements, misrepresentations and erroneous corporate disclosures concerning AI, intellectual property infringement, algorithmic bias, and even employee sabotage. And more so than with traditional risks, AI risks are inherently unique to each business because businesses will utilize the vast array of AI technologies in myriad different ways and to different degrees. These risks and challenges are moving from the hypothetical to the real world.

A lawsuit filed last month in the U.S. District Court for the Northern District of California highlights how these risks are materializing and impacting businesses. In that lawsuit, three authors on behalf of a putative class allege that AI developer Anthropic infringed on their copyrights by using their copyrighted works to train its models. This legal action follows hundreds of other lawsuits related to the use of AI.

Insurance & Indemnification: A Holistic Approach to Risk Management

Given the broad spectrum of AI-related risks, it is essential for businesses to thoroughly assess their insurance coverage. Traditional policies like errors & omissions (E&O), cybersecurity, commercial general liability (CGL), and commercial property might offer some protection, but they do not address the full spectrum of risks stemming from AI. Similarly, while certain insurers are starting to offer specialized AI-specific insurance products, the insurance market has not yet developed insurance products that fully insure the wide array of potential AI risks.

This is where indemnification agreements come in. While the insurance industry sorts out the precise scope of coverage available under legacy and AI-specific insurance products, indemnification agreements can temporarily fill risk management gaps. By including indemnification provisions in contracts, businesses can clarify responsibilities and reduce the likelihood of disputes over liability when losses occur. Indeed, businesses like Microsoft, OpenAI, Google, Adobe, and Getty Images have done just that by incorporating some form of indemnification in certain AI-related contracts.

Using indemnification agreements as another tool in AI risk management toolkits offers several potential strategic advantages. Perhaps most importantly, indemnification agreements can be drafted and negotiated to cover specific scenarios uniquely likely to materialize in the AI context. This high level of customization allows businesses to plug specific risk holes left by their existing insurance programs.  

Indemnification agreements can not only effectively plug coverage gaps, but they have potential benefits in their own right. To start, indemnification agreements may provide parties clarity about their obligations and reduce the likelihood of legal disputes. Indemnification agreements may also help deter reckless behavior. That is, if a party knows that it will be held financially responsible, it may act more carefully to avoid triggering indemnification obligations. Indemnification agreements may further foster stronger partnerships between businesses. When two parties sign an agreement that includes indemnification clauses, it may signify an understanding of the risks involved and a shared commitment to addressing them. That shared commitment can build trust and collaboration.

Conclusion

As AI continues to reshape the economy, businesses must adapt their risk management strategies to address the unique challenges presented. Insurance will remain a cornerstone of these strategies, but it does not have to be the only tool in the toolbox—especially when the insurance marketplace does not yet offer solutions for the full set of possible AI risks. By incorporating indemnification agreements into their risk management plans, businesses can better protect themselves from the multifaceted risks associated with AI.