What the hell is artificial superintelligence?
Sounds sorta… silly, honestly. Yeah, we need a course correction on AI, or more, we need a significant course correction on big tech. But a city council ‘banning’ it, without understanding / clarity on what they’re banning, is nuts.
Like here they define a bit clearer what they’re banning –
ban autonomous systems that can modify their own core algorithms, run code outside their initial sandbox or perform complex tasks without continuous human oversight if those tasks could affect municipal networks, banking, financial markets or public safety.
Most enterprise firewalls would fall into this category, email filters too. Got an email system that automatically updates its algorithm to block malicious attackers? Banned. Even a simple script to help with sorting through financial data – like doing money-laundering/fraud checks overnight, with an outputed report in the morning for staff to investigate when they get in, would potentially fit the ban criteria.
Most 24/7 continuous monitoring systems that use algorithms to send alerts/notify humans would fall into this category – they’re specifically designed to watch for complex variable combinations, and then alert humans when potential issues shows up, so they’re explicitly designed not to be running “continuously under human oversight”.
“AI” is such a broad/undefined term, that blanket banning tends to also ban technology based on algorithms that’ve been around for multiple decades, without significant issues. Sorta like how the EU’s initial AI regulations had to make all these absurd carve outs for things like “Credit Scores”, because a credit score is technically an algorithm, and they were legislating algorithms.
The ban’s also nearly impossible to enforce, given its scope, and where “AI” has its hooks. Like if you ban AI, but all of Microsoft’s products have it entrenched at this point… while I don’t agree with it, I guarantee you the government / council would rather be in a state of non-compliance with their own law, rather than use Linux.
Where do we file applications for the position of city Bladerunner?
With Artificial Super Intelligence now banned, Sam Altman announced plans to direct more r&d resources towards Artificial General Intelligence instead.
People are talking about the ban itself but in what fucking world would a law only be repealed with consent of “95% of registered voters” LMAO
What does that mean?
What does this law actually ban? ChatGPT? I think the fuck not.
Does it ban something that doesn’t exist yet, but might exist someday? If so, what is it, other than a feel-good measure that does nothing?
Apple is putting a tiny LLM on your phone.
I hope this means the “i” in “iPhone” isn’t for “indiana”.
Unwanted AI on your smartphone while driving through town? That’s a fine and a jail sentence. Genius.
Don’t call it artificial superintellgience FFS ☠️☠️☠️
Why not? It sounds hilariously stupid (which is accurate).
It’s really easy to make the argument an LLM is not an “artificial super intelligence” because it’s not. Remember it’s the letter of the law that often matters not the spirit of it
But what does the law do when the people making it call it that as well? Seems like a really easy argument to make, pointing at press releases, tweets by CEOs, IPO filings etc. using that language?
It’s a real industry term, not a riff off Trump’s dumbass SI.
LLMs are not artificial super intelligence! They’re just text prediction machines with some reinforcement learning and are not smart enough to be considered such
It’s a fake industry
Ughh… don’t call it superintelligence. That’s not a thing.
bigly intelligence
ASI is an industry term. This is not Trump’s idiot SI.
So you’ll say anything the “industry” says? Even though it’s fake?
Nobody fucking cares, bootlicker
The people pushing it the most are anti-intellectuals.
We need to call it fake intelligence.
That’s what it’s already called, “artificial” is a synonym for “fake” in most contexts.
Right, but artificial is a four syllable word, clearly beyond their grasp.
the word artificial presents an an artificial dichotomy between the “natural” world and the “manmade” world. because we are neither animals nor part of the environment.
So, artificial?
Yes - maybe I should have been clear about what I was implying. Artificial is clearly too big of a word for them to understand.
Dont blame your goof on other people being stupid, own it.
It’s a thing in the sense that unicorns are a thing. There aren’t any currently, but it’s a coherent concept. If they banned unicorns within city limits, the ban wouldn’t do anything unless someone managed to make a unicorn; same idea with ASI.
It’s a real term.
“Artificial Superintelligence (ASI) is a hypothetical stage of AI where a computer system surpasses the cognitive performance and problem-solving abilities of the most gifted human minds across every domain of interest.”
- that’s an “industry term” in the same way that “panopticon” is a prison industry term, or that “brain in a vat” is a neurology term. It’s a name made up by science fiction writers and philosophers to describe a thing that does not exist, and has never existed. Might it exist in the future? Maybe, but that’s irrelevant to an issue of law. You can’t pass a law banning an ill-defined “panopticon”, because that isn’t a term of art enshrined in any US code.
- because, as you say, it is purely hypothetical, such a law would be unenforceable, and would likely fail a test in court.
- because it is, once again, nonexistent as of yet, the burden is on you to prove that they aren’t specifically referring to the term which the Shit lord of Hamberder enshrined in his executive declaration, because that use of “Superintelligence” actually has a basis in current US government parlance.
I really want to see how they measure the “cognitive performance” of an LLM…
I also want to know what part of problem solving exists outside of these tests, as it’s actually a core component of cognitive testing.
Seems like someone made shit up, I’m guessing using an LLM…
that’s not a thing
I suspect this is the point
you don’t have to change the law if you change what words mean, and in this case, they did not ban llm’s. they banned a thing that does not exist
In the US, you cannot ban people from using an app on their personal phone in your community. This is just ignorant. I’m not an AI bro, but you can’t just have each community deciding on what constitutional rights they want to violate.
They’re only banning super intelligence. That doesn’t exist as a product.
Well didn’t trump rename AI into SI?
Not sure freedom to use apps is included in the US Constitution …
That would likely violate the first amendment. Many of the ways you use your phone are considered forms of speech/expression. Wholesale banning of sectors of these by a government is a legitimate violation of your rights.
Fuck AI, tho.
Tell that to all the places that banned porn.
Originally porn was protected by the first amendment.
There’s a whole movie about this called the people versus Larry flynt that kind of explains why.
Well, you see, something something God.
My god can beat up your god. Their name is “God That Beats Up Other Gods.” Therefore you have to do what I say. I even got a note to that effect
Think of the children!
Religious Leader Zealots: “Yes, of course we are.”
AI is something that violates every posible law (moral and legal). It should be hard banned.
Yes. Every. Possible. Law.
AI raped my mom.
AI quartered troops in my home.
In all seriousness, if someone wanted to sue that local government referring the first amendment, they would likely take the ban down.
That said, I don’t find any actual article from a reputable news outlet describing what actually happened and what is in the city council ordinance. Likely a brief unenforceable article, which, as such, has no bearing on anyone’s life.
A redditor mentioned they banned “self improving” ai, which, if true, would explain why it isn’t getting more coverage- this is a nothing burger.
A redditor mentioned they banned “self improving” ai, which, if true, would explain why it isn’t getting more coverage- this is a nothing burger.
The year is 2029, Meta’s GrokBots have taken over the world. They approach Bloomington Indiana, their next target. A kindly older gentleman approaches the army on the outskirts of town.
“Sorry boys, we banned self improving AI back in 2026, you are not allowed here.”
The robots nod and continue onward, parting around the outskirts of town like a river flowing around a rock.
The town is saved. It becomes the last bastion of humanity within 6 months.
a big sign outside town says
: ;; / | / | .' : .-' ' _.-' / .-*" / _ .-' .' _.-*?' .' .' .-" .' __ .' , .-' .-+. .' _.-*".' / \ .-' _.--**""**-. .-' _.y-:-" .' : `+. .*""*. `. :-. -. \ .' ; .--*""*--. / __ ` _.--. \ |$| -.` -.;/ _.-+. : .' :*" "*..*" y`-' $| ;*" _( \ / +----/ / .'.-'---+ .-._.+' `. -'_.--. :- "_( `*-: | \/\/\/ | /) ` .'___ ' "_( ; `._| robots.txt | \ )` .'.' `./_" ( : \| | (`._..--**" : .- ; `"' \ | | `----**"T"" " `+. | `. | Fuck off | ' .' : _.-*"*- | godsdamn | / / ' .-*" _ | toasters | __..-'\ / "+,'___..--| |--**"" `-.__.' "" +----------------+
It’s basically just writing a letter with extra steps
The output technically belongs to the AI company, does the first amendment apply to non human entities? If it does I don’t think anyone should be supporting that application …
Banning it outright is ableist, there’s no need include the right to use certain apps in the Constitution.
There aren’t disabilities that require access to the plagiarism machine; it’s insulting to act like disabled people can’t write or make art.
there are few things ablebodied people find more offensive than a disabled person who can outperform them.
It’s a product. It can be banned. It’s dumb and shows a general lack of understanding by those in charge but I don’t see what this has to do with the first amendment or the constitution more broadly.
And a product that doesn’t exist!
Sounds like § 307 StGB
Is that the nuke one?
Edit: lokked it up, yes, it is.
This is so dumb and exactly what I would imagine from Indiana. Everyone in here cheering it on is braindead.
Life is never as simple as people like this wish it was.
The last step is laziness born of becoming jaded. Its easier to generalise, so people tell themselves it was the answer all along. Stupid trap, dont fall for it.
I think having a realistic perspective can be instrumental to making good decisions. A lot of our strategy historically has hinged on the idea that almost everyone is good, and that’s evidently not true.
Describing people as broadly good is both true and a useless generalisation, hard times can still push good people to do bad things.
Fast forward a hundred years and this place will be the last refuge from the incursion due to archaic laws written a century prior.
Men with sticks never legally allowed to buy guns, famously winners of the next war.
How would this be enforced?
Easy, if a company in Bloomington says “it was just our AI chatbot,” busted. If a person says “well, Grok told me,” busted - hope you live outside city limits. If a small business uses the slop font, send in the SWAT team, we got 'em boys. Tune all the city’s Flock cameras to watch everyone’s phones and we’ll root out this cancer!
You’ve missed the entire point. ASI is a real industry term and it does not exist, may not exist in our lifetimes. This was a safe bet by the council to make them look like they’re doing something about AI when all they did was to ban leprechauns.
And… no leprechauns to be found. So, mission accomplished, remember to vote!
I’m voting for the no-leprechauns party! They really know how to get things done.
Mass surveillance, which is why it’s misguided for anyone to be happy about this
This is supposedly the actual ordinance
However it’s not published on their official government site (yet?), which only has up to ordinance 28 published (the anti_AG one is #29). I can’t find any indication this has actually passed (note that the doc linked above isn’t signed and has an empty space for date-of-passage). But maybe the site just isn’t updated.
https://webmail.bloomington.in.gov/council/legislation/Ordinance/2026
I guess the interesting parts are, the justification for the law
Under Indiana Code § 36-1-3 (The Home Rule Act), the Common Council has sovereign authority to regulate public safety hazard nuisances within its jurisdiction.
The definition of what’s banned, which, as other comments have pointed out, doesn’t actually exist
Superintelligence” shall be analogous to an Artificial General Intelligence (“AGI”) with capabilities of Agency (A), Generality (G) and Intelligence (I), wherein an AGI does possess super-human capability in all three aspects, of AGI, as an AGI wherein any such autonomous Agentic (A) system capable of self-modifying its core algorithms, executing code in environments outside its initial sandbox, or operating without continuous human oversight while Generally (G) performing any complex goal-oriented task that may affect municipal infrastructure, financial markets, or public safety, in combination with an extreme level of super-human Intelligence (I), and defined by the AI Industry standard terminology “Superintelligence” herein defined as “Any intellect that greatly exceeds the cognitive performance of humans in virtually all domains of interest.”
And that it doesn’t say you can’t use it, it says you can’t “deploy” or “distribute” it
No person or corporation shall distribute or permit to be distributed nor deployed into the premises of any citizen or business in our City, any Superintelligence for any purpose. Deployment, for the purposes of this ordinance, shall include the remote execution, activation, or utilization of any unauthorized digital agent through a locally accessed interface, browser, API, or connected hardware within the City of Bloomington, regardless of the physical location of the server infrastructure hosting the system.
However it’s not published on their official government site (yet?), which only has up to ordinance 28 published (the anti_AG one is #29).
50k Satoshi says that nobody knows how to update the site without AI assistance, and the AI is preventing them from doing that.
LLMs don’t have agency or intelligence so this does nothing
