What’s Self Got to do With It?

The October edition of Best’s Review contained an article with the headline, “Self-Driving Shift Comes With Early Costs, Changing Landscape for Insurers”. All things considered, that may be an understatement. Here are three reasons why, among others:

Exhibit A

Although AVs [autonomous vehicles] haven’t fully hit the road, insurers are already getting a glimpse of the future—including … a legal liability shift.

AVs can move fault from the driver to the manufacturer, the software developer, or the AI system due to defects in design, hardware, or algorithms, leading to increased products liability claims. Accidents may involve shared fault among owners, operators, fleets, and suppliers, complicating claims allocation and potentially resulting in higher payouts or subrogation against manufacturers. Employers or fleet operators could face vicarious liability for business-use accidents, while drivers may be negligent if they fail to override the system when required.

Exhibit B

You’re starting to get a glimpse of what repair costs could look like for these kinds of vehicles, and where that complicates liability assessment … while frequency may go down, severity can increase, influenced by the higher costs of repairs with these vehicles having more advanced parts and technology required … To achieve cost reduction, a decline in frequency would have to meaningfully offset the potentially higher severity from repairs.

Using telematics for real-time monitoring of AV performance and crash data to refine risk profiles and detect transition risks during adoption will complicate underwriting. Claims adjudication will shift from eyewitness accounts to analyzing complex technical data, demanding new tools like AI for incident reconstruction, and collaborations with experts in engineering and AI. Higher repair costs for AV tech, shorter vehicle lifecycles, and the need for dynamic add operational challenges. Expensive AV components like LiDAR sensors raise repair and replacement costs, necessitating higher policy limits.

Exhibit C

A major question of accident liability received its first answer in August when Tesla was ordered to pay more than $240 million in a fatal crash involving a car that used its Autopilot software technology.

Translation: We ain’t seen nothin’ yet.

We can’t know everything. But as AVs proliferate, we can be sure insurers will have to offer flexible policies that cover manual and autonomous modes, with endorsements for electronic equipment, data loss, and cyber-physical risks that aren’t yet addressed in standard forms. Commercial fleets will require elevated auto liability and physical damage coverage, often bundled with technology E&O coverage.

Beyond traditional auto, insurers will have to consider cyber liability for hacks, D&O coverage for executives, and usage-based insurance tied to AV data. They’ll also have to consider silent cyber risks like physical damage from digital attacks and avoid as yet unforeseen issues in which intentional AI actions trigger exclusions. And they’ll have to allow for varying premium levels depending on the levels of autonomy in particular vehicles, fleet usage, and telematics data, with on-demand adjustments for switches between manual and self-driving modes.

The Moral of the Story

The moral of this story is that the core processing vendors who serve those insurers, will have to ensure the platforms evolve to handle the changes. Legacy systems aren’t equipped to handle data-intensive, AI-driven workflows or to manage uncertainty flexibly.

When it comes to self-driving vehicles, self has much to do with it.