A host of insurance policies could have grounds to deny liability following an AI-related healthcare incident, medical indemnity underwriter Tego warns.

CEO Eric Lowenstein says a potential “stack gap” means cyber, technology errors and omissions, medical malpractice and product liability policy terms could all dispute where responsibility lies.

“The hospital’s medical malpractice carrier says the AI made the call, not a clinician. Cyber insurers say no network intrusion, no data breach – not our line. The vendor’s tech E&O excludes bodily injury. The vendor’s product liability is in dispute over whether AI is even a ‘product’ in that jurisdiction,” he said.

“Every policy has a plausible reason to decline. The hospital is sitting underneath all of it. That is the gap.”

Tego is developing a standalone artificial intelligence liability product aimed at Australian technology providers and organisations deploying AI systems.

Mr Lowenstein told delegates at the Kennedys Law conference in London this month that insurance was “built for a world where machines did not make decisions. That world is gone. It is the first time our industry has had to ask not just who is at fault, but whether the concept of fault still applies.

“The new chain is patient, clinician, hospital as deployer, AI vendor, foundation model provider (LLM). Each link is a potential defendant. Each link points at the next one when something goes wrong.”

A systemic insurance risk is posed by an AI application “monoculture” in which a small number of foundation large language models are supplied by just a handful of global providers, Mr Lowenstein says. All fraud detection systems in banking, customer service agents in retail and AI scribes in healthcare sit on top of just a few foundation models.

“A single defect in one widely used foundation model could trigger claims across thousands of unrelated insureds simultaneously, across every industry, across countries, across policy lines. That is a fundamentally different shape of risk to anything our industry has priced before.”

He also warns of a “look-back problem”, in which organisations retrospectively analyse historical records using new AI tools, potentially finding fresh liability exposures.

“The act of looking is itself what creates the liability,” he said.