Top
Best
New

Posted by theahura 6 hours ago

There is a shadow hanging over this Fable thing(12gramsofcarbon.com)
300 points | 267 commentspage 2
duffydotsvg 45 minutes ago|
Disagree with the popular conclusions that this is either calculated anti-marketing or politically-motivated axe grinding. Both theories might carry some weight, but they belie the bigger picture. If you believe AI will be a determinant of economic growth, defense capacity, scientific advancement, and geopolitical supremacy over the next few decades, then state supervision/control was always inevitable. There's already precedent in nuclear tech, chip tech, aerospace tech, etc. In theory, the stakes here are 100x higher. USA and China will do whatever they can to get a competitive edge over the other. Meaning this is probably just the first salvo in an ongoing series of similar events. At the end of the day, I'm not sure any of it will matter. All signs point to the models being basically impossible to contain.
zkmon 57 minutes ago||
> run a ton of agents in parallel most of the time

What makes you think everyone (and government) should play along and align with your way of dependency on AI? Not even 1% of the people use AI the way you do. Fable model is not a basic need. Government represents the average Joe. You could also say "I make a ton of nuke weapons and this government has stopped the public sharing of how to make them!".

Havoc 56 minutes ago||
>This was announced on 5:21 PM on a Friday. Sorta a suspicious time. Whenever someone does something intentionally on a Friday evening, my first thought is ‘o, the markets.’

The insider trading kleptocrat has found a new toy

2gremlin181 5 hours ago||
Potentially relevant: https://news.ycombinator.com/item?id=38361050
agnosticmantis 4 hours ago||
Counterintuitively, this is a huge win for misAnthropic and other closed labs in the US. They can nerf the models, ask for IDs from users and do what it takes to comply with whatever regulation they've been fighting for.

Foreign labs releasing open source models won't be able to comply, and as a result open source models will remain stunted at pre-mythos levels or their use will be criminalized.

We should look past the petty fights these closed labs have, and see their common interest in banning open source and/or local models.

dpe82 3 hours ago|
Why would foreign (relative to the US) models suddenly sit still? There's enormous incentive to improve; surely they'll be able to figure out how just like their American counterparts?

We've seen this movie before with crypto export bans in the 90s. The rest of the world caught up and then surpassed the US very quickly - and that was without the enormous financial incentives of AI.

sbmthakur 3 hours ago||
Even without those incentives, the messaging is clear: you don't want your inference to be shut arbitrarily. Export controls are nothing new but a lot of people have underestimated them due to globalization and the general nature of software. This is a good opportunity for entities around the world to get their setup going.
matt3210 5 hours ago||
What a coincidence, Anthropic getting handicapped so xAI can try to catch up
johnwheeler 5 hours ago|
XAI rents out compute to anthropic. I feel like Sam Altman is behind this that little rat.
maxbond 5 hours ago|||
What makes that more likely than that people at the DoD are alarmed (with or without good justification) at Fable's capabilities plus finding a jailbreak (or what they interpret as a jailbreak while Anthropic seems to dispute the requests met the level Fable should refuse)?
rustcleaner 3 hours ago||||
Considering he cornered future production of DRAM, I believe it!

Profitability rests on subscribers. Never subscribe!

cm2187 4 hours ago|||
How is that going to help him? "Our models are so inferior they are not deemed a threat unlike anthropic's"?

I think it is either a missile directed at anthropic, as retribution for not giving the DoD what it wants, in which case it is likely to resolve pretty quickly. Or it is a shift of policy toward export restrictions on powerful LLM and then every model will be impacted as they reach the threshold. In which case this could have massive implications of revenues, valuations, and the whole datacenter buildout. And frankly on the location of the white collar workforce if it is indeed a productivity multiplier, all countries reciprocate, and not all countries can match the US LLMs.

And why would the EU allow exports of chip manufacturing equipment if the US then restricts the export of derivatives of those chips to the EU?

llelouch 4 hours ago|||
There are many people in this administration invested in OpenAI like the Kushner's. They are attacking Anthropic however they can. You will notice a lot of propaganda on social media sites against anthropic. It's very obvious.
thepasch 1 hour ago||||
> How is that going to help him? "Our models are so inferior they are not deemed a threat unlike anthropic's"?

Do you honestly think that this - logic and reason - is going to stop anyone from hyping whatever nonsense he comes up with to the moon and back anyway? Right after the SpaceX IPO of all things?

trhway 4 hours ago|||
>How is that going to help him?

the first one to do IPO will win big. With the government pressing Anthropic, OpenAI IPO will vacuum up the funds that otherwise would have went into Anthropic IPO as OpenAI was falling behind.

>and the whole datacenter buildout.

somebody just did a $2T IPO with the idea of datacenters in space. One can wonder what laws/jurisdiction those datacenters will be subject to.

airport_barfly 4 hours ago||
Everyone's focusing on marketing and market manipulation here, but the real consequences are more serious IMO.

If a volatile administration can ban you from running code that you wrote -- without any democratic processes like a law or lawsuit -- why would you build anything in the US?

tedggh 3 hours ago||
With Anthropic history of using the news as their free marketing agency, I remain a bit skeptical. My guess is that something will be worked out in the next hours or days and Fable will be back.
smooc 4 hours ago||
This should be a red herring for Europe (and others using US models).

Every non-American company is now at a disadvantage against American companies. The implications can not be overstated.

graemep 3 hours ago||
Google Deepmind is headquartered in the UK and has R & D in multiple countries. How would the US ban non-nationals from using something that is largely developed outside the US by people who are not US nationals?

Mistral might be a bit behind but this might give them a lot more business.

Most of all, a lot more people will switch to Chinese models. They will catchup, soon enough.

I have not had much of a chance to try Fable, but it did not seem better than Opus for what I tried it out on. Maybe its better on bigger jobs/vibe coding type tasks which is not something I do anyway.

cbg0 2 hours ago||
The US can create sanctions with legal repercussions, the same way they've done with Iran or Cuba in the past.
graemep 2 hours ago||
Certainly to stop people with US businesses from doing business with Iran or Cuba. To stop an American owned British company from doing business in the UK? To stop a French company doing business in France? To stop either doing business with the rest of the world? To stop anyone doing business with China?

Its really not comparable.

cbg0 2 hours ago||
They absolutely can impose sanctions on foreign companies by restricting their access to US markets, investments and penalizing US banks doing business with them.

For some EU companies this is irrelevant, but for global companies this becomes a problem.

graemep 36 minutes ago||
So did they stop any Iranian companies doing business with Iran? Did they stop China or Russia doing business with Iran? If Google Deepmind has to stop doing business with non-US citizens what will they do with their R & D in the UK and other countries? Will Mistral stop doing business in France? Will the Chinese AI companies stop doing business with everyone else (including China?) to retain access to US markets and banking?

The likely end result of this is that it will shrink the market to which American companies have access by more than it will shrink the market for anyone else.

sajithdilshan 3 hours ago|||
It indeed is a wake up call. But at the same time the strong data protection laws, copyright and privacy laws make it extremely difficult for a European company to develop a frontier model. Activist lawyers can sue and drag startups for training their model on a news article and the legal expenses would be higher than the engineering costs.

ChatGPT was released 4 years ago and still out of 27 countries in EU, only Mistral based in France has a model closer to a frontiers and IMO EU has already lost the race and still trying to catch up to yesterday models.

dpe82 3 hours ago||
As we learned with export restrictions on crypto in the 90s, that disadvantage will be short-lived and backfire in the long-run.
JimsonYang 5 hours ago|
I seriously feel like there's easier ways for OpenAI to catch up to anthropic and it would be a waste of political capital that the idea of Sam pulling strings for this to happen seems highly unlikely
More comments...