Merged Artificial Intelligence

[Frank Newgent]

https://apnews.com/article/antitrust-lawsuit-ai-slowdown-anthropic-openai-spacexai-google-960af4308161eaf4ed13c383b0ce1c1b

anti trust suit claims big ai companies illegally colluded to slow ai development

Would AI companies together agreeing to strengthen safety and security processes violate antitrust law? Weird.

Guess the logic here is that governments must step in and do that job.

 

[Wudang]

Ads in chatGPT don’t seem to perform well. Gosh.

ChatGPT ads: media agencies are ‘underwhelmed’

In February, OpenAI introduced advertising in ChatGPT. Ad buyers love this idea: [AdNews] ChatGPT reaches people who have effectively opted out of traditional advertising. Digitally native, ad-bloc…

pivot-to-ai.com

 

[arthwollipot]

I've provided two examples already.

Crucially, how long ago did those happen?

 

[Roger Ramjets]

That's why those same billionaire tech bros are advocating for such a regulatory regime? To save everyone else from themselves? Of course not. They're being selfish bastards, they aren't trying to protect you.

In business everybody's a selfish bastard.

Nevertheless, Elon Musk has long been warning about the potential dangers of AI, and now Sam Altman is joining him. I think both of them are genuinely concerned about what the industry might accidentally unleash.

This is not unheard of. The Health and Morals of Apprentices Act 1802 (AKA Factories Act) was introduced by factory owner Sir Robert Peel, who said:-

Having other pursuits, it was not often in my power to visit the factories, but whenever such visits were made, I was struck with the uniform appearance of bad health, and, in many cases, stinted growth of the children; the hours of labour were regulated by the interest of the overseer, whose remuneration depending on the quantity of the work done, he was often induced to make the poor children work excessive hours, and to stop their complaints by trifling bribes. Finding our own factories under such management, and learning that the like practices prevailed in other parts of the kingdom where similar machinery was in use...

Nobody should be more aware of the potential danger from advanced AI than the visionaries who are creating it, and we shouldn't dismiss their warnings just because they stand to profit from it. Even if their concern is completely self-serving (which I don't think is true) they still want AI to be seen in a positive light, and so have a vested interest in preventing it from going rogue.

China is backing US groups opposed to AI and data centers. Did you not know this?

That's not what I asked.

China AI Regulation Overview

Interim Measures for the Administration of Generative AI Services (2023): Cross-sectoral administrative framework governing public-facing generative AI services, requiring lawful training data sourcing, content moderation, risk mitigation, labeling, and algorithm registration...
Interim Measures for the Administration of Anthropomorphic AI Interaction Services (2025/2026): Interim regulation governing AI simulating human personality and emotion, mandating AI identification prompts, anti-addiction controls, minor and elderly protections, and safety assessments.
Ethics-Safety Guidelines for Artificial Intelligence Applications 1.0 (2026): TC260 technical guideline articulating nine ethics-safety principles, operational risk controls, audit logging, continuous model reviews, and multidisciplinary safety boards.

The answer is, China already has a strong regulatory framework in place for AI.

As for opposition to data centers (which is not the same thing as opposing AI):-
The theory taking the rich by storm: China funds data center haters

The theory that China is paying local activists in America to oppose data centers is catching on like wildfire among the Silicon Valley elite, despite a lack of evidence. Wealthy investors are naming names and pointing fingers, even if they can't directly link their projects' opponents to China... 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."

Now we see the reason China is being blamed for opposition to AI data centers, to dismiss the opposition from real Americans who don't want it in their back yard.

 

[Ziggurat]

The answer is, China already has a strong regulatory framework in place for AI.

That remains to be seen. China has lots of regulations that are not actually well enforced. About the only rules they do consistently enforce are those that directly protect the regime.

 

[Roger Ramjets]

No kidding. A human runs at about the same power requirement as an old lightbulb, that is, 100W. Unless they're doing some heavy physical work or exercise, of course.

Tesla's Hardware 3 FSD AI processor uses a maximum of 72 watts. With this it's able to drive as well as a human. HW4 uses up to 144 watts, and is about 5 times better than the average human driver.

There is no reason believe AI won't become dramatically more power-efficient in the future. The only reason it hasn't happened yet is that electricity is cheap and abundant. However this only applies to data centers, not eg. cars and robots. Future AI architecture will be quite different from the LLMs of today. It will work with smaller data sets and be much smarter. This will happen because there is a big incentive to do it.

 

[Gawdzilla Sama]

Let's consider AI's grandkid, Skynet.