VHS [he/him]

This paper looks like it’s essentially run by the US weapons industry. It’s a subsidiary of GovExec who helps military contractors sell to the government. This means they have an interest in drumming up armed conflict and you’d have to take everything they print with this in mind.

A group called Cofacts allows users to forward dubious messages to a chatbot. Human editors check the messages, enter them into a database, and get back to the user with a verdict.

“This year, because gen AI is just so mailable [sic], they just fine-tuned a language module together that can clarify such disinformation…adding back a context and things like that. So we’re no longer outnumbered,” she said.

I’d be highly skeptical of how well this would work. A language model is not a truth machine, it’s just a program that tries to write like a human. There are countless examples of ChatGPT and similar programs saying blatantly false things and making up citations out of thin air.

*removed externally hosted image*

i feel bad for any writer there who’s just trying to do tech journalism

cnet’s record is terrible. first they do crapware bundling, then have factually incorrect articles written by machine learning algorithms, now this

there are definitely real concerns to be had about addictive and manipulative software, i use linux and a dumbphone as well. but “AI” is not some genius, it’s a dumb algorithm and i think ascribing these apocalyptic powers to it is ridiculous.

and using the word “consoomer” makes you look like a 4chan nazi whether that’s what you are or not.