Remix clone Hacker News

Well I mean, nitpick, but Fentanyl is a useful medication in the right context. It's not inherently evil.

I think my biggest concern with AI is its biggest proponents have the least wisdom imaginable. I'm deeply concerned that our technocrats are running full speed at AGI with like zero plan for what happens if it "disrupts" 50% of jobs in a shockingly short period of time, or worse outcomes (theres some evidence the new tariff policies were generated with LLMs.. its probably already making policy. But it could be worse. What happens when bad actors start using these things to intentionally gaslight the population?)

But I actually think AI (not AGI) as an assistant can be helpful.

▲

Terr_ 20 hours ago | parent | next [-]

> I think my biggest concern with AI is its biggest proponents have the least wisdom imaginable. [...] (not AGI)

Speaking of Wisdom and a different "AGI", I think there's an old Dungeons and Dragons joke that can be reworked here:

Intelligence is knowing than an LLM uses vector embeddings of tokens.

Wisdom is knowing LLMs shouldn't be used for business rules.

▲

brain5ide 20 hours ago | parent | prev | next [-]

Are we talking about structural things or about individual perspective things?

At individual perspective - AI is useful as a helper to achieve your generative tasks. I'd argue against analytic tasks, but YMMV.

At the societal perspective, e.g. you as individual can not trus anything the society has produced, because it's likely some AI generated bullshit.

Some time ago, if you were not trusting a source, you could build your understanding by evaluating a plurality of sources and perspectives and get to the answer in a statistical manner. Now every possible argument can be stretched in any possible dimension and your ability to build a conclusion has been ripped away.

▲

walterbell 20 hours ago | parent [-]

> build your understanding by evaluating a plurality of sources and perspectives and get to the answer in a statistical manner

A few thousand years of pre-LLM primary sources remain available for evaluation by humans and LLMs.

▲

coryrc 19 hours ago | parent | next [-]

You and I remember pre-AI famous works. "Hey, I'm pretty sure Odysseus took a long time to get home". Somebody goes and prints 50 different AI-generated versions of the _Odyssey_, how are future generations supposed to know which is real and which is fake?

▲

walterbell 18 hours ago | parent | next [-]

> how are future generations supposed to know which is real

Reality/truth/history has always been an expensive pursuit in the face of evolving pollutants.

	▲	coryrc 5 hours ago \| parent [-]
		That's definitely true. History has been thoroughly manufactured by humans. Naively, I thought the storage of computers might preserve first-hand accounts forever; it might, but it might not be discernible.

▲

noosphr 19 hours ago | parent | prev [-]

This is literally how the Odyssey was passed down for the 2000 years before the printing press was invented.

Every work had multiple versions. All versions were different. Some versions were diametrically opposed to others.

Have a look at Bible scholarship to see just _how_ divergent texts can become by nothing more than scribe errors.

▲

coryrc 5 hours ago | parent | next [-]

They were real because they were made by people all along. Now you can't tell.

I think you're right my analogy is imperfect. I'm only human (or am I? :P)

▲

samtheprogram 19 hours ago | parent | prev [-]

99.9999999% sure that was their point? Why else would they bring up that particular work?

▲

burnished 6 hours ago | parent [-]

Because they thought it was an ancient and unchanging text.

	▲	coryrc 5 hours ago \| parent [-]
		No, but it was a bad example because I was thinking only of the authorship point of view. A better example would have been the complaint tablet to Ea-nāṣir. We're pretty sure it's real; there might still be people alive that remember it being discovered. But in a hundred years, people with gen AI have created museums of fake artifacts but plausible, can future people be sure? A good fraction of the US population today believes wildly untrue things about events happening in real time!

▲

namaria 18 hours ago | parent | prev [-]

I know how to swim yet a riptide can still pull me out to sea

	▲	walterbell 18 hours ago \| parent [-]
		Symbols can help, https://www.weather.gov/images/safety/rip/Rip_Currents_Sign4...

▲

spooky_action 17 hours ago | parent | prev | next [-]

What evidence is there that tarrif policy was LLM generated?

▲

calcifer 16 hours ago | parent | next [-]

There are uninhabited islands on the list.

▲

KoolKat23 15 hours ago | parent [-]

Despite people's ridicule this is normal practice, prevents loopholes being exploited.

▲

mr_toad 14 hours ago | parent [-]

It seems more likely that bad data was involved.

There are actually export statistics (obviously errors, possibly fraud) for these islands. Someone probably stuck the numbers in a formula without digging a little deeper.

	▲	Izkata 8 hours ago \| parent \| next [-]
		Yep, there is data saying those islands are exporting to the US: https://www.theguardian.com/australia-news/2025/apr/04/revea...
	▲	KoolKat23 12 hours ago \| parent \| prev [-]
		Well in this case it's also beneficial given how automated supply chains are. It's probably the most sane aspect of the whole thing.

▲

af78 16 hours ago | parent | prev [-]

There are people who asked several AI engines (ChatGPT, Grok etc.) “what should the tariff policy be to bring the trade balance to zero?” (quoting from memory) an the answer was the formula used by the Trump administration. If I find the references I will post them as a follow-up.

Russia, North Korea and handful of other countries were spared, likely because they sided with the US and Russia at the UN General Assembly on Feb 24 of this year, in voting against “Advancing a comprehensive, just and lasting peace in Ukraine.” https://digitallibrary.un.org/record/4076672

EDIT: Found it: https://nitter.net/krishnanrohit/status/1907587352157106292

Also discussed here: https://www.latintimes.com/trump-accused-using-chatgpt-creat...

The theory was first floated by Destiny, a popular political commentator. He accused the administration of using ChatGPT to calculate the tariffs the U.S. is charged by other countries, "which is why the tariffs make absolutely no fucking sense."

"They're simply dividing the trade deficit we have with a country with our imports from that country, or using 10%, whichever is greater," Destiny, who goes by @TheOmniLiberal on X, shared in a post on Wednesday.

> I think they asked ChatGPT to calculate the tariffs from other countries, which is why the tariffs make absolutely no fucking sense.

> They're simply dividing the trade deficit we have with a country with our imports from that country, or using 10%, whichever is greater. https://t.co/Rc45V7qxHl pic.twitter.com/SUu2syKbHS

> — Destiny | Steven Bonnell II (@TheOmniLiberal) April 2, 2025

He attached a screenshot of his exchange with the AI bot. He started by asking ChatGPT, "What would be an easy way to calculate the tariffs that should be imposed on other countries so that the US is on even-playing fields when it comes to trade deficit? Set minimum at 10%."

"To calculate tariffs that help level the playing field in terms of trade deficits (with a minimum tariff of 10%), you can use a proportional tariff formula based on the trade deficit with each country. The idea is to impose higher tariffs on countries with which the U.S. has larger trade deficits, thus incentivizing more balanced trade," the bot responded, along with a formula to use.

John Aravosis, an influencer with a background in law and journalism, shared a TikTok video that then outlined how each tariff was calculated; by essentially taking the U.S. trade deficit with the country divided by the total imports from that country to the U.S.

"Guys, they're setting U.S. trade policy based on a bad ChatGPT question that got it totally wrong. That's how we're doing trade war with the world," Aravosis proclaimed before adding the stock market is "totally crashing."

▲

XorNot 18 hours ago | parent | prev [-]

Honestly this post seems like misplaced wisdom to me: your concern is the development of AGI displacing jobs and not the numerous reliability problems with the analytic use of AI tools in particular the overestimate of LLM capabilities because they're good at writing pretty prose?

If we were headed straight to the AGI era then hey, problem solved - intelligent general machines which can advance towards solutions in a coherent if not human like fashion is one thing but that's not what AI is today.

AI today is enormously unreliable and very limited in a dangerous way - namely it looks more capable then it is.