Insurers are beginning to address the risks associated with artificial intelligence by offering new insurance products for companies facing potential losses from AI-related errors. Lloyd’s of London has introduced a policy through the startup Armilla, which covers legal claims against businesses if customers are harmed by malfunctioning AI systems. As reported by the Financial Times, this initiative comes in response to increasing concerns over mistakes made by AI tools, such as customer service chatbots that can produce misleading or false information. The consequences of relying on inaccurate AI outputs can be severe, leading to financial losses and reputational damage for companies. Kelwin Fernandes, CEO of NILG.AI, emphasized the accountability dilemma, questioning who is liable when a human transfers responsibility to an AI system. Recent incidents, such as Virgin Money’s apology for its chatbot’s inappropriate response and Air Canada’s legal troubles over a fabricated discount, highlight the urgent need for such insurance products in the evolving landscape of AI technology.
While I’m talking AI, I wanted to note something. With the selection of a new Pope, CNN highlights how the new pontiff selected his name as a nod to a previous Pope Leo, who guided the church during the industrial revolution. ““In our own day, the church offers to everyone the treasury of her social teaching in response to another industrial revolution and to developments in the field of artificial intelligence that pose new challenges for the defense of human dignity, justice and labor,” the new American pontiff said Saturday, speaking in fluent Italian.”
Why do we care?
Once insurers are willing to underwrite specific AI risks—particularly errors leading to financial or reputational harm—it signals formal recognition that AI systems are operational risk vectors, not just “smart tools.” his elevates AI failures from IT incidents to insured business events. Companies offering AI services or operating AI-based solutions (especially client-facing ones like chatbots or recommendation engines) now need to consider how to structure SLAs, disclosures, and fail-safes to meet underwriter requirements.
While the Lloyd’s policy is high-profile, it’s still niche and largely untested. The scope of what it covers, exclusions, and how claims would be assessed in practice is unclear. Insurance policies may offer a false sense of security without solving the upstream governance issues. While the Pope’s AI framing is notable (and why I highlighted it), it doesn’t have direct impact on tech strategy. However, it does reflect a mood—and that mood could influence future social policy. But tying it too tightly to business trends risks overreading the moment. Notable, however.