Our vulnerability is an open internet that includes Russia, leaving our populace completely open to Russian influence campaigns of micro targeted propaganda, recruitment, and incitement.
I like how the “Chinese propaganda” against datacenters was found to be mostly a collection of facts.
While Russian propaganda is more clumsy and emotional, we’ve produced such an unequal society with such low regular education and outright corruption that people fall for it. That says something negative about US, not them.
If we admit our system cant even survive an open internet, it should be dismantled or reformed rather than free exchange closed.
I lived in a communistic country, where BBC/Voice of America/Radio free Europe were banned/jammed, under the pretext of 'your feeble minds will not be able to resist the insidious allure of western burgeois propaganda'.
Seeing the same argumentation from the west's side is disheartening. We are now in the west, but (almost) full circle back where were before.
Education would blunt, but not eliminate the threat of nation-state funded micro targeted influence campaigns. Moreover, it would take years, and we have the problem now.
In terms of data center propaganda and fact. Both are true. China is funding anti data center sentiment for geopolitical power, while at the same time, there are legitimate reasons for people to not like/ or fear AI. The problem is we (the U.S.) have to solve both. We cannot let China have significantly better AI models than we have because that is a real national security threat, while AI also threatens potential mass destruction dangers and an avenue of utter concentration of wealth, etc
I’m not buying the “AI security threat”, and I don’t really care if the best models are open source - and Chinese. If everyone can buy enough hardware to run the best models, what’s the threat? The same tired arguments as banning books on encryption or chemistry?
I see the crazed AI investment as worthless financial gambling, a house of cards that will eventually crumble, and the argument that “we need to fund AI” really is “your tax dollars should bail out our financial mistake, because security”.
It’s nonsense all the way down. Covid showed we didn’t even manufacture basic drugs in the country anymore, and were openly threatening our biggest/best/longest allies (Canada, Mexico, NATO). AI is so far down the security list as to make it noise, compared to our recent missteps.
Linvill said he can't rule out that the Chinese are quietly paying influencers or others to oppose data centers, but he pointed out there's no need.
"I think in this case, the people talking about data centers are real people with real passions and real perspectives and real opinions."
In Utah, those people appear to have forced Kevin O'Leary to scale back his plans. After the president of the Utah Senate raised concerns, he agreed to shrink his data center project to a quarter of its original size.
—— End NPR article quote
So yeah, no need to produce Chinese anti-datacenter propaganda - people have valid reasons, and they hate it. It’s already blocking and scaling back projects.
Oh they absolutely are valid, though! We've seen them ignore permitting for generators, use incredible amounts of water and raise all our power bills, steal books and movies for training - and for what?
Subscription AI isn't going to pay for itself - it's pretty clear that they'll want a bailout in the name of "security"; which is why it's so great the Chinese are releasing models as-good-or-better for free: ironically because it drives an even bigger stake into the heart of US AI companies.
The sooner these AI companies are destroyed, the better.
You seem to know much more about this topic than I do, and your assessment agrees with mine. Predicting rare and sudden events is a very difficult modeling problem generally, and the data limitations is real. I agree that it will probably need to come from coupling of ML with physical simulators plus more extensive 3D map data on force vectors, material properties, etc.
Where is this money coming from?
They can strike a deal, but the money has to come from somewhere and if it came from money directed at wind energy, then this will be illegal, I'd venture to guess.
Yeah it's a good question, but unfortunately large infrastructure projects require a lot more than just formal legal protections of their funding and permitting.
If it takes just one retard at the top to disrupt a massive decades-long project, even in a way that eventually gets negated by a court, then ta-da, we just don't get any more big projects.
The courts, eventually-probably. What this does to the business environment and future endeavours regarding clean energy, however, is an exercise left to the reader
As a scientist, I can confirm. Reality will likely kick their assess. Intelligence and creativity is not the bottleneck. Great scientists have 50 good ideas for every one they actually manage to execute on the grant->experiment->manuscript haul.
What if the grants process were AI's on your side understanding what you want to do talking to the AI of the funding source, and they could provide immediate and direct feedback on what they would or would not fund and why? And it just runs one a week, looking for new funding sources and applying if relevant.