Recent searches
Search options
Alleged #X whistleblower wrote a thing about #Elon's election manipulation that sounds mostly plausible to me (see replies)
> "One of the most disturbing things we did was create thousands of fake accounts using advanced #AI systems called #Grok and #Eliza. These accounts looked completely real and pushed political messages that spread like wildfire. Havn't you noticed they all disappeared? Like magic"
> "We knowingly allowed foreign governments to manipulate the platform. State-backed groups from #Israel, #Iran, and #Russia were running their own influence campaigns"
> #ElonMusk himself was deeply involved in these decisions, often joking about being "Black Hat #MAGA."
> "What started as US #electioninterference has now spread to other countries. We're doing the same thing in #Germany and other #European nations. The damage we've done is immeasurable, and I don't know if it can ever be fixed"
https://theconcernedbird.substack.com/p/elon-musks-and-xs-role-in-2024-election
Plausibility should, however, not be sufficient to believe that something is true. There's not enough there, there yet.
@eloquence agree, though a) the stuff about #ElizaOS is compelling and b) i've seen other credible people prior to this talking about Elon using bots to pump #TSLA on Twitter. i may have even seen the proof.
@eloquence also remember this chart showing the inflection in right wing engagement on the platform the day Elon endorsed trump?
It's plausible, but nothing that Eliza does is fundamentally novel; character.ai has been in that business since 2022, including with similar "character definitions". I created some myself when the product came out. There have apparently been 20K chats with my "Carl Sagan" bot since I created it. :)
Do I think that Elon did more than what's known, which is already bad enough? Absolutely. But I'd have to see more evidence for the specific claims by Mr. Bird to believe them.
@eloquence yeah we're on the same page. i think i'm also conflating Eliza (the bot) with #ElizaOS (which IIRC was just released to the public a few weeks ago - which fits the pattern better than Eliza).
https://universeodon.com/@cryptadamist/113921141658152549
FWIW though Saudi Arabia's bot/sock puppet network is well documented at this point. I've also spoken to reporters at The Times Of Israel who've explained #Netanyahu's vast bot network to me, so I don't doubt anything about letting foreign governments run influence operations. I mean we know Elon eliminated the entire Twitter staff so at a minimum no one was guarding the hen house and I was talking to the Times of Israel and BBC about this like back in early 2023.
@eloquence also also i witnessed the rise of Mario Nawfal's account first hand which Elon was obviously involved in (he literally flew his plane to Mario's studio, which was insane).
Mario's account's rise has been proven beyond a shadow of a doubt to have been driven by bots. There's tons of discussion about it on the bird app - his engagement was just transparently fake.
@eloquence in fact the only think that the alleged whistleblower says that doesn't match the pretty well fleshed out fact pattern i've already confirmed is the disinfo stuff. Mario Nawfal, at least, doesn't spread blatant #disinfo. But his account is like, the huge one. Seems highly likely there's smaller accounts doing far more dastardly things. I mean, for one thing all the bots that boost Mario are constantly boosting crypto scams.
There's lots we can separately say and demonstrate, including about how he's built up other accounts like End Wokeness, Autism Capital; boosted his own visibility; used the "verification" economy as a way to boost the right, etc.
It's just that I could easily create a "whistleblower account" with an LLM that looks even more plausible than Mr. or Ms. Bird's.
@eloquence yeah i hear that. my initial take was actually to not believe that it was a real whistleblower even though I believe almost everything written in the substack post. feels a bit PR stunty, like maybe someone who knows these things to be true but didn't actually work at X wanted to get the word out.
but either way Elon didn't build up #AutismCapital (p.s. fuck that guy) the same way Mario Nawfal was built up. not only are they very different kinds of accounts, AC was much more organic whereas Nawfal was wholly synthetic and is constantly showing up at RT's offices or whatever the other Russian propaganda outlet is called. Mario posts photos, even, of himself in Russian state media offices.
@eloquence also i think it's a lot more than just paid verification that boosted the right... the Queensland U study chart haunts me.
Yeah, very good point. Following @timothyjgraham now.
@eloquence @timothyjgraham thx for the tip; followed.
this is the #ElizaOS i was conflating with #Eliza. came out a few weeks ago and enables one to orchestrate a fleet of AI chatbots. was created by memecoin buttpumpers who IIRC don't claim to have had any #MarcAndreessen / #a16z funding (but don't claim to not have had such a thing either and the launch of ElizaOS was meteoric).
@eloquence @timothyjgraham and actually come to think of it in 2024 it would be pretty dumb for a whistleblower NOT to have an AI write the whistleblow message text... having AI compose the text will strip the prose of any identifying characteristics (misspellings, incorrect pronoun use that would be common for an H1B Hindi or Mandarin speaker, abbreviations, etc).
same with ransom notes - going forward all that stuff will be written by AI