Insurance policies for companies that cover financial and reputational damage resulting from undetected bias in AI algorithms, despite using monitoring tools. This provides a safety net, but also incentivizes superficial compliance.