It’s easy to criticize car co’s for programming vehicles to break traffic and parking laws. But human drivers do it constantly.

Two big differences:
🔹 Humans make context-specific decisions (i.e., is it an emergency?). Software doesn’t.
🔹 Humans are liable for their actions. Car co's often aren’t.

Will add (3) people are also largely able to understand & mitigate the violations of other people because they fall into a shared framework.

Those are often regional, like "speeding +10 is OK" or "everyone here double parks". Algorithms do not share in those predictable frameworks.

Colombia, Guyana, and Brazil must be very surprised

TBF they clarified that it is only a "subset" of users who have catastrophic problems.

Which then makes me wonder how much they know about sets.

In Japan, South Korea, Taiwan, Hong Kong, Singapore… masking is boring. Etiquette. Courtesy.
A cough? Pollution day? Crowded underground? Mask goes on. End of story.

No identity crisis. No tribal signalling. Just collective care. /2
(Family photo from SE Asia - man in mask=NBD)

This may sound strange, but many students either consciously or subconsciously do not see school (including college) as an opportunity to "learn." The transactional nature of the system is totalizing. Even students who don't want to outsource their learning to LLMs feel like maybe they should...

And so: "Our support for innovation and America’s leadership in AI does not extend to using our residents, especially children, as guinea pigs while AI companies experiment with new applications. Nor is our support for innovation an excuse for noncompliance with our laws." Boom.