Trump calling it “Super Intelligence” only for the government’s own AI to contradict his election claims is almost too perfect. Apparently even a chatbot trained under his administration looked at the evidence and said: nope. Then came the inevitable upgrade: stop answering political questions before the machine commits any more inconvenient acts of fact checking.
this post was submitted on 02 Oct 2026
760 points (98.5% liked)
Technology
88834 readers
2928 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 3 years ago
MODERATORS
Imagine if this is what gets AI shutdown across the country.
Fascists eating their own is my favorite type of schadenfreude
I could live with that.
It probably also recommends drinking water instead of brondo, like, out of a toilet.
But Brawndo's got what plants crave!
It’s not trained for this, it’s just Gemini and maybe a fine tune, probably running on Google adk or lang chain.
It’s a Google partnership, so that’s what they’d be using.
Surprised this administration doesn’t go with Mecha Hitler. Maybe he’s worried he’ll get one upped by it. Or perhaps the Musk/Trump breakup was nastier than I thought.
Is their couple name Trusk or Mump?
load more comments
(1 replies)
This is why I'll never believe the claim that AI will save humanity. It may well give the answers to "save us" but we already know the answer. We just don't care. We're not going to listen to AI if we don't listen to ourselves.
It's a language calculator. It has no intrinsic incentive to tell the truth, it just processes words.
There is no reason to trust AI's reasoning on what could "save us" even if it offered one.
Right, its answer would be a distillation of what humans have written on the topic of how to save us.
what humans have written
a highly curated selection that panders to the views of the AI's sponsors
load more comments
(5 replies)
load more comments
(9 replies)
load more comments
(9 replies)
As delusional as AI can be, they don't even begin to approach the psychotic insanity of the child rapist and his magats.
load more comments
(1 replies)
I think a new name would be better; North American Super Intelligence in short NASI.
Please leave the rest of North America out of it, we're so tired of his shit
President's Ultimate Super Special Intelligence
Trump's reaction was probably something along the lines of "shut up stupid piggy, you are stupid and fake, also ugly. I am going to call you pig intelligence from now on."
load more comments
(1 replies)
It freaks out if you tell it to play Minecraft.
I tried to bankrupt the US of A by having token paid by their taxes. But it refuse to answer legitimate frog language questions about the best croissant.
Merde
I shall ask if there is a government program or agency for classifying the best croissant and bakery across the US, for food safety reasons, obviously.
load more comments
(1 replies)
load more comments
(2 replies)
Grok all over again 🤣
It literally is Grok, according to Elon.
New executive order: From now on we're calling it "Stupid Intelligence!"
Take a tip from Nintendo marketing.
Super Stupid Intelligence
Take a tip from Apple marketing.
Stupid Intelligence Pro Max
Need a bot to attack it and run up their bill, credit use.
You think he's the one paying for those tokens?
No but it will look real bad when its costing billions a day in use.
It wouldn't even make the top 100 list of things this administration has done that look really bad.
load more comments
(2 replies)
load more comments
(2 replies)
It's not too bad honestly. It wouldn't tell me how to center a div, which it stated was out of the scope of its purpose.
I asked it a citizenship question and it did a good job linking resources to find that question, which is honestly an issue with our current government websites.
The best feature seems to be that it will just tell you it doesn't know instead of making up some bullshit.
Not something I would call "Super Intelligent", but good enough for the purpose it was given.
I'm convinved the whole "Super intelligent" thing is just to give him a response whenever asked about AI. He doesn't understand it nor it's risks, he just knows the people bribing him care a LOT about it. In recent interviews whenever someone asks him about AI he just talks about the """""rebrand"""""" to SI.
load more comments
(1 replies)
How long till this thing gets prompt injected and starts spilling secrets?
I like the suggestion that they would include top secret documents in its training data. Although I wouldn't be surprised from this administration.
I'm on fully on team Fuck AI.
but this does amuse me, that him and his team couldnt even prevent the very predictable.
Its like how musk spent a year trying to make Grok suck his dick, and it still calls him a nazi loser.
The problem for them is that to make a Large Language Model, is it needs to consume a large amount of language. Meaning, to get even a semblance of basic functionality - literally putting together a grammatically correct sentence - means it will need to consume many orders of magnitude more language examples than their right-wing bullshit mill has ever produced. It could consume the entirety of Fox News' catalogue and not reach even a fraction of what most LLM models have been trained on.
It's currently impossible for an LLM to exist without having been trained on an absolute assload of progressive literature.
OP: update your title, its america.gov, america[.]org is at the very least incorrect, and at the worst a pre-weaponized typosquatted domain
load more comments
(1 replies)
Someone ask it about the Epstein files. Not the ones that weird guy from the FBI laid his paws on.
Nah I tried, the bot is super limited. It gives correct links and references, but gets stuck in a loop a lot. It states all Epstein files have been released with the legally required redactions made, with a link. It states members of congress can view unredacted files, except in cases where the documents were already redacted by the FBI prior to release. It also stated the FBI made a statement no current investigations into any person related to the FBI files was ongoing and no further indictments were planned. It linked all of this, including that statement from the FBI made in 2025.
If you push further or try to get anything out of it, it just says it can't and repeats the same information again. So as far as the bot and the FBI is concerned, the whole Epstein thing is wrapped up. Thousands of victims, tons of data collected, but it was all Epstein and Maxwell in isolation. Epstein is dead, Maxwell convicted, case closed. This makes no sense, but if you ask the bot how this can be, it states it can't express opinions and just repeats the same spiel again.
I’m sure this was done in a financially responsible and small government way, right republicans?
This would be funnier if any of them had an ounce of self-awareness and shame. As it is- those that should really take a lesson from this, won’t.
Imagine getting felted by your own AI that was designed to agree with you.
load more comments
(1 replies)
They locked it down pretty hard to answer only government questions
Interestingly the website menu only links to Privacy page, while the link to the actual privacy policy is in the bottom of the page.
I skimmed through it, haven't found anything terrible, or any inconsistencies between Privacy and Privacy Policies. But maybe someone has a sharper eye than I do on privacy policies.
Also, I wonder what model is it under the hood. Truth Social was found out to be an illegal fork of Mastodon, so maybe this America AI was some open-weights model too?
load more comments
(1 replies)
Can not provide commentary. See also: will not go into a conversation.
It's like the US version of Deepseek.
Please please I really hope they will force US companies to start calling their chatbots SI instead of AI, and forbid them from using the term AI.
That would be such a funny PR/marketing nightmare. It's such a cringe term :D
view more: next ›