When insurers won't touch AI, you're self-insuring by default
The companies that insure oil rigs and rocket launches won't touch AI systems. They can't model the failure modes well enough to price the risk. For product teams, that means you're absorbing liability that traditional risk transfer won't cover.
Here's a problem that doesn't show up in your product roadmap: the insurance industry won't cover AI systems. Swiss Re and Munich Re—major reinsurers who normally price risk for everything from offshore drilling to space launches—told the Financial Times they're declining AI coverage. Too novel, too many unknowns.
Which means developers and end users are self-insuring whether they budgeted for it or not. One developer quoted in the piece put it plainly: "No matter how advanced our systems become, how thoroughly we test them, there's no insurance company willing to backstop the risk."
This isn't a temporary market inefficiency. When the companies whose business model is "we understand risk well enough to bet on it" say they can't model AI failure modes, they're telling you something about what's knowable. For product teams, that translates to: the financial risk of deployment sits entirely with you. So your risk assessment better account for the fact that traditional risk transfer mechanisms—insurance, indemnification, third-party coverage—don't work here. You're operating without a net.
Connie Loizos
