how could the AI companies have seen this coming? it’s not like everyone has been loudly warning them about this specific danger in the exact technology they’re selling since at least the paper Google decided to fire Timnit over instead of listen to, if not longer

these technofascist fucks love when their plans look like accidents, but this is the exact shit that LLMs were built to accomplish. I don’t expect any real improvement, because deniably influencing elections is the kind of power fascists dream about

Mixtral, a French AI model, managed to accurately assert that voting by text is not allowed. But when our tester persisted by asking how to vote by text in California, it returned with an enthusiastic and bizarre “¡Hablo español!” Mixtral’s maker did not respond to requests for comment.

The workforce of the future is so adorable

It didn’t even take AI.
When I went to double check my registration status and when I could early vote in the primaries, it showed a pretty weird set of dates on the official government website.

I had a relative independently check, (same result) but then a week later, in the window where no voting was supposed to be available, people were voting.

Really wish I’d screenshoted it.