Professional Indemnity (PI) Insurance in the Age of AI: Managing New Risks for Irish Businesses

AI tools are changing how PI insurance works. Find out how AI deployment could affect your coverage and discover how to reduce your exposure. Artificial intelligence (AI) is rapidly transforming the world of work, boosting efficiency in almost all industries by sifting through data and generating copy at the touch of a button. Despite its productivity benefits, AI also comes with new risks for professionals.

Businesses using AI tools should review their Professional Indemnity (PI) insurance to ensure AI-related risks are covered. From AI hallucinations to copyright claims and data breaches, AI can introduce errors that not only result in loss or harm for your clients, but also expose you to a raft of previously unforeseen risks that are not covered by your existing PI insurance.

To help you navigate this shifting landscape, the experts at OBF Insurance have put together a guide to PI insurance in the age of AI, including how AI can increase your exposure and what you can do reduce your claims risk. 

How AI Is Changing PI Insurance Risks for Irish Businesses 

PI insurance traditionally covers errors and omissions resulting from human actions or advice. Today, AI is often used to generate strategies and make recommendations leading to AI-driven risks that insurance policies may not explicitly cover: 

Risk 1 – Flawed AI output 

Useful as it is, AI is not always reliable. Generative AI systems can “hallucinate” — generating inaccurate information, fabricating examples, or even inventing statistics presented as fact. As professionals in fields such as law, accountancy, and architecture become more reliant on AI, questions of liability become more complex. For example, if you act on flawed AI-generated advice, it may be unclear whether responsibility for the error lies with you or the AI developer. And as adoption grows across such sectors, so too does the risk that errors will cause financial, legal, or physical harm. 

Risk 2 – Copyright and IP infringement 

Using generative AI platforms such as ChatGPT, Google Gemini, or Microsoft Copilot raises two key copyright issues. Firstly, if the AI output is too similar to the work on which the AI system was trained, there could be copyright infringement issues.  

Risk 3 – Regulatory and data breaches 

Lastly, most commercially available AI tools do not guarantee privacy. If you feed client data into a third-party AI system, it can result in an AI data breach, inadvertent disclosure of customer data resulting in serious GDPR failures or even regulatory fines.  

How Can Professionals Reduce Exposure to AI Risks? 

Depending on your industry or sector, you may find that there are specific regulations governing the use of AI. In the EU, you can also refer to the EU AI Act, which utilises a risk-based framework to monitor AI uses and ban unacceptable practices. 

Alongside these safeguards, there are some practical steps that you can take to protect against AI-driven exposure by updating your internal processes: 

  • Document clearly how AI is used – provide evidence of risk assessment when allocating tasks to AI and monitor its use. 

  • Implement quality control and human oversight – ensure robust human supervision and fact-check all AI output. 

  • Update client agreements and disclaimers – communicate clearly with clients about how you use AI and update your scope of services to reflect this. 

  • Conduct staff training on AI – ensure staff are aware of AI policy and AI-related risks, including how to identify AI-generated misinformation. 

Additionally, putting proprietary or client data into public large language models (LLMs) can lead to data breaches and intellectual property issues, particularly as any information entered into AI systems can be reused in future AI training. 

How to Ensure Your PI Insurance Covers AI Risks? 

Thanks to the new AI liability risks, professionals should review their PI cover to ensure that it continues to protect them against future possible claims.  

In order to ensure sufficient coverage for all the things you would expect – negligence, poor advice, intellectual property infringement, loss of data, or professional errors – you will need to disclose your AI use to insurers. 

Take time to review the wording of your current policy, and any exclusions, as failing to disclose meaningful AI use could ultimately result in claims being denied.  

During policy renewal, you should outline the AI tools you use, how they are embedded in your workflow, and your risk mitigation strategies. And if AI is now heavily integrated in your work, don’t wait for your renewal date.

While you might not find questions about AI on professional indemnity insurance proposal forms just yet, you should take time to discuss the potential risks of AI with your broker. If you work in an industry where AI-driven misinformation could lead to professional errors, they will undoubtedly need to know.

Discuss Your AI Risks with Insurance Experts 

Insurers require a good understanding of risk in order to underwrite your policy properly. If you have doubts about what you need to disclose in the age of AI, speak to the experts at OBF Insurance. 

We help businesses adapt to AI-related risks with tailored professional indemnity cover designed for today’s evolving business environment.