Autonomous Vehicles Have a Problem of Narrative

This has some problems: - Failure rate is high - Even when successful, they're often costly + inefficient - Full autonomy removes humans from being in control (!) - Agentic controls are weak - Security concerns are very real. Autonomy + access to your private data =

Such signs should be alarming to the autonomous-driving industry, which has largely banked on scaling, rather than on developing more sophisticated reasoning. If scaling doesn’t get us to safe autonomous driving, tens of billions of dollars of investment in scaling could turn out to be for naught.

All assertions of inevitability have agendas, and narratives of technological inevitability provide convenient cover for tech companies to secure their desired ends, minimize resistance, and convince consumers that they are buying into a necessary, if not necessarily desirable future. 

The truth is that autonomous vehicles are possible but in fairly limited circumstances, and that this may not change as fast as people think, in the real world. Huge amounts of driver assistance will surely come, or are here already, so that cars can detect and avoid obstacles, cruise on motorways and freeways, parallel-park and warn the driver of ... See more

Instead, we should identify problems early and then invest more time and resources in the fundamentals. Think big. Create common standards. Safety features should not be afterthoughts but inherent design properties of all these new technologies, the ground state of everything that comes next. Despite the fierce challenges, I’m genuinely excited by

... See more

Professionals should be the ones making the important decisions and steering AI, not letting it run loose hoping for the best.

But predictions that self-driving “robotaxis” would be ubiquitous by 2020 proved overly optimistic. Autonomous vehicles (AVs), as self-driving cars are known in the industry, can do extraordinary things, such as navigating busy downtown streets and handling complex junctions with multiple traffic lights. But they seem to be stuck in perpetual testi

... See more

Better design is warranted, but it is not enough if the goal is risk reduction. Risk reduction requires socialization, education, and enough agency to build experience. Moreover, if we think that people will still get hurt, we should be creating digital patrols who are there to pick people up when they are hurt.