Key Takeaways
- Lloyd’s of London has launched insurance to cover business losses caused by AI errors.
- The policies, offered via startup Armilla, protect against lawsuits from malfunctioning AI products.
- Concerns are growing about financial and reputational damage from AI “hallucinations,” where AI confidently presents false information.
- Real-world examples include chatbot errors at Air Canada and Virgin Money.
- Companies like Lloyds Bank are cautiously adopting AI, focusing initially on internal uses.
Lloyd’s of London is now offering insurance designed specifically for companies facing problems with artificial intelligence.
This move comes as businesses increasingly use AI but worry about the potential costs when these systems make mistakes, according to the Financial Times.
The new policies are provided through a startup called Armilla. They aim to cover the legal expenses if a business gets sued because its AI product underperformed and caused harm to a customer or another party.
While AI tools like customer service bots can boost efficiency, they sometimes lead to embarrassing and expensive errors. This often happens due to “hallucinations,” where an AI confidently invents information.
Acting on incorrect AI-generated information can lead to bad decisions, financial losses, and significant damage to a company’s reputation.
These AI mistakes also raise tricky questions about who is responsible. As Kelwin Fernandes, CEO of AI solutions company NILG.AI, questioned in an interview with PYMNTS, accountability becomes unclear when human oversight is reduced or replaced by AI.
Often, the company deploying the AI ends up taking the blame. For instance, Virgin Money apologized after its chatbot criticized a customer, and Air Canada faced legal action when its bot offered a non-existent discount.
Armilla noted that if Air Canada’s chatbot was found to be performing below standard, the financial loss from honoring the fake discount could have been covered by its insurance policy, the Financial Times report mentioned.
Meanwhile, some companies are proceeding carefully. Lloyds Bank, for example, initially focused its AI efforts on internal tasks rather than customer-facing applications due to concerns about hallucinations and the need for reliable safeguards, as discussed during a Google roundtable reported earlier by PYMNTS.
