Comment on Grok can't apologize. Grok isn't sentient. So why do headlines keep saying it did?
spit_evil_olive_tips@beehaw.org 5 days agoReuters is the worst offender that I’m aware of. they sneakily changed their headline and rewrote the article:
Elon Musk’s Grok AI floods X with sexualized photos of women and minors
but luckily someone archived it, with the original title:
Grok says safeguard lapses led to images of ‘minors in minimal clothing’ on X
(and you can still see that original headline in the URL of the Reuters link above)
besides the headline, that original article is only 7 short paragraphs and contains 4 “Grok said…” and a “Grok gave no further details” - it’s not just quoting Grok like it’s a real person, it’s only quoting Grok and no one else.
and almost as infuriating as the “Grok said” shit, the Reuters headline also repeated the fucking disgusting “minors in minimal clothing” euphemism that Grok itself used in its “statement”.
Segab@beehaw.org 4 days ago
Journalists love using AI as a source because it’s a creative way of using the passive voice which sounds more neutral. Instead of attributing an action to a person like “Elon Musk’s Website Used for Generating CSAM” historically they’d use something passive, impersonal and vague like “Vulgal Pictures Are Being Generated On X”, but of you’re relatively informed you see read past the phrasing and picture how Musk is accountable. Now journalists can use the active voice so it appears more tangible but pin in on AI to sound neutral and hold nobody accountable, which serves the status quo.