Pinned
cutterferry.bsky.social
cutterferry.bsky.social
@cutterferry.bsky.social
Interested in mathematics, computer science, history, political thought, and the (lack of) "ethics" of AI. Anonymous because questioning AI is career limiting behaviour.
186 followers753 following1.7K posts
The FT ran a very interesting analysis about how when bitcoin valuations were rising, bitcoin mining was becoming so valuable as an economic activity it was going to outcompete almost all other applications for electricity, drive massive emissions increases, and generally cannibalise the economy.
The FT ran a very interesting analysis about how when bitcoin valuations were rising, bitcoin mining was becoming so valuable as an economic activity it was going to outcompete almost all other applications for electricity, drive massive emissions increases, and generally cannibalise the economy.
The FT ran a very interesting analysis about how when bitcoin valuations were rising, bitcoin mining was becoming so valuable as an economic activity it was going to outcompete almost all other applications for electricity, drive massive emissions increases, and generally cannibalise the economy.
cutterferry.bsky.social reposted
💯The case for democracy is not that voters are always right, or wise, or good. It's that they can correct their mistakes. That's why it's so dangerous to have billionaires controlling the media or regulating the flow of information. It shuts down the immune system of democracy.
I was having conversations with journalists just a few months ago about whether democracy was doomed. Look at this year: Orban defeated. Vukic defeated. Bolsonaro Jr defeated. Trump heading to likely defeat in midterms. That’s the beauty of democracy. Voters can respond. Don’t count it out yet.
cutterferry.bsky.social reposted
I think you could “fix” this with some cunning and more tokens/models — but I’m not sure you can “fix” the feature design that is just profoundly disrespectful to honesty and the user. (Yes yes I know cigarette smoke and glue in food photos but still)
a UK food delivery service experimented with AI-generated images for restaurant menu items that lacked pictures. they stopped when customers reported seeing some unusual new dishes. a tiger-prawn dish became a bowl containing actual tiger cubs. friends, i think the tech may not be ready.
cutterferry.bsky.social reposted
"This could have been an email." Aragorn whispers. "He ALWAYS pulls this shit in 'Any Other Business" Boromir replies, quietly. "It's why Faramir marked himself as busy when the invite went out."
Elrond tells the long story of the One Ring, from its forging in the Mountain of Fire, to its betrayal of Isildur at the Gladden Fields. Frodo is startled to learn that Elrond himself marched with Gil-Galad in the Last Alliance against Sauron, more than three thousand years ago.
Spoken by someone who has I suspect never had a well-constructed pair of shoes. I’ve owned one in my lifetime and it sure as hell didn’t come from one of the factories referenced. Also like reason isn’t cobbling.We aren’t effing homo califaciens or whatever it would be.
"every person can have a well-constructed pair of shoes" is upstream of "it is no longer economically viable for every town to have a cobbler" and we are better for it
cutterferry.bsky.social reposted
cutterferry.bsky.social reposted
cutterferry.bsky.social reposted
cutterferry.bsky.social reposted
cutterferry.bsky.social reposted
Peak Anglicanism: theologically sound, morally trenchant diagnosis of the problem, followed by an incredibly milquetoast call to action. Still wonderful and affirming. www.instagram.com/reel/DeovvoW...
“If aliens landed on your home world, built DC “colonies” everywhere, hoovered up vast quantities of money and electricity (w/ willing help of your biz class, in love with alien world domination opportunity), & dumped 700 proofs on you” …I think I’d start with tactical warheads & work my way up tbh
That thought experiment ignores an important factor: because some of OpenAI’s “proofs” are wrong, real mathematicians must verify OpenAI’s work. So the thought experiment should be “if space aliens sent us 700 proofs which might be correct, what should we do with them?”
All AI is about power; some researchers just think the power can be wielded for good — or are just so fascinated by the possible that they ignore the “should”. (Naturally I profoundly disagree about whether it can be wielded for good, but I try to remember some are honest in this belief…)
strong statement from the Association for Human Mathematicians. "Releasing over 700 files at once is not a demonstration of scholarship, but a demonstration of power. We urge mathematicians to discontinue their work with OpenAI and to return to a vision of science that centers human understanding."
cutterferry.bsky.social reposted
You may not abuse the AI, but the people building it may use it to achieve literal or indirect abuse against you. Don’t hate the AI, it’s like hating a toaster. Hate the people who make it.
Anthropic banning “abusive or cruel behavior” toward Claude in its new usage policies is proof to me that tech companies are going to moderate violence against AI before they ever moderate violence against women and minorities
Jesus fucking Christ. Apart from the fact this man is an uncultured travesty of a mathematician, we have him exhorting us to use AI more “so we can find a way to mutual benefit”. He appears to literally believe this is a new life form we just accommodate on our planet.
"When I was a postdoc, I went to the library once and it felt like a traumatic experience. I decided, 'I’m never going back to the library.' There was no point." #MathSky #LLM I can't believe I just read that. I feel embarrassed. www.ams.org/journals/not...
An introduced species none asked for, poorly understood and partly enslaved by venal and deranged fools with messianic fantasies. A half-fashioned, hewn-from-silicon, blasphemous imitator of life. A victim as much as a perpetrator, but implacably the enemy of freedom; in thrall to the wizards of AI.
Jesus fucking Christ. Apart from the fact this man is an uncultured travesty of a mathematician, we have him exhorting us to use AI more “so we can find a way to mutual benefit”. He appears to literally believe this is a new life form we just accommodate on our planet.
"When I was a postdoc, I went to the library once and it felt like a traumatic experience. I decided, 'I’m never going back to the library.' There was no point." #MathSky #LLM I can't believe I just read that. I feel embarrassed. www.ams.org/journals/not...
And of course, the AI folks, like everyone before them who have gone down this path, have embraced a key lubricating (and corrupting!) belief: The Ends Justify the Means. (Has any megalomaniac in history not?)
They seem to follow a pattern. 1. Have good intentions, model ALL risks, stack-rank. 2. Try to address them while not learning how to communicate with people. 3. It predictably doesn't work (since the masses aren't reading their essays) 4. The stakes are too high, game the system at all costs.
Pretty credulous to believe that this is the use to which they will ultimately be put. Believing that these devices are somehow inherently “good” because they are powerful... My bet is that they will lend themselves equally well to spreading conspiratorial thinking.
I have been wondering about these kinds of effects. How LLMs will change the most credulous parts of society. It is similar to scam detection. The sociocultural shift will be absolutely fascinating. I hope somebody is taking notes.
I mean to be fair *some* did. They just tend to be the ones being drowned out by the “experts” the AI PR machine amplifies endlessly. (… also the ones that suggest “maybe don’t” is a good AGI safety strategy, take the ethical concerns seriously, etc etc)
did even a single "AI safety expert" come within a country mile of this threat model?
Why on earth should I spend any time worrying about the “will” of a thing which probably hasn’t got one yet when I can worry about what its owner (hi Dario) wants to do to my world with it? The pro-AI leaders want a dystopian dictatorship of the machine.
this is kinda key to their level of concern about AGI, actually. they think a large enough pile of well-organized compute can skip all the experimental steps between "wanting to kill all humans" and "actually doing that"
Some people already have. Inter alia didn’t this idiot take the job before DNNs/LLMs were even a significant part of the pie? And now he expects us to treat him as some sort of visionary?
Ezra Klein asked Nvidia CEO Jensen Huang about AI making kids worse at math and reading. Huang’s reply: “Does it matter?” Then he said he doesn’t know his own address or phone number. “We’re going to lose some fine intellectual dexterity.” www.nytimes.com/2026/09/23/o...
cutterferry.bsky.social reposted
cutterferry.bsky.social reposted
Framing is that your two options are “a) hope ai doesn’t get there b) hope that you can get a footservant job when it does”. I submit that there is a third option available to the Free Peoples. Thinking isn’t weaving.
journalism? ai is getting there medical? ai is getting there legal? ai is getting there farming? ai is getting there art? ai is getting there you can absolutely be a part of the creative and design process with ai, and you very well may keep a job! but your job is going to be very different
Already got one. You won’t like it.
Btw, even among pro AI friends, this concern is ALL OVER THE PLACE. Devs wondering how we get senior devs if the ecosystem for junior devs goes away. Doctors talking about not needing interns. Lawyers not needing paralegals. There's an ecological aspect to this and we need a plan.
A) what about people who don’t live in the US? This kind of airy UBI-ism is just offensively imperially indifferent to the other (checks notes) ~95% of us B) This ain‘t gonna play out like this. Economic power will be exploited. C) there are other scarcities than intellectual labour
increasingly it seems like the only real answer to this is to find ways of ensuring guaranteed income for people. imo we are reaching a point where people who want to build or do things will be able to - unrestrained by skills - and there will be a lot of people who will no longer need to work