Artificial Intelligence thread

magmunta

Junior Member
Registered Member
Pornography and sex toys are two different things. Government focused on regulating pornographic text, images, and videos than silicone.

So I think it's unlikely that erotic chatbots will be legalized.
Shouldn't be legalized under a competent government. Young people are vulnerable to sham emotional attachment, delusions and not acquiring basic social skills. Should be regarded as a free market failure such as gambling or light drugs. Maybe it is even a better idea to impose age restriction similar to age restrictions in gambling.
 
Last edited:

Wrought

Captain
Registered Member
Harvard piece on the Chinese structural advantages w.r.t. agent integration.

China’s edge here is not in models. It is in plumbing. Five conditions that rarely coexist happen to line up in the same market: permission infrastructure, execution capacity, ecosystem orchestration, consumer readiness, and regulatory sequencing. Payments, identity, and authorization are so deeply embedded in daily life, through Alipay and WeChat Pay, that once a user grants permission, agents can execute without repeated handoffs. China’s dense logistics and on-demand service networks, especially in food delivery and local services, reliably convert digital intent into real-world outcomes. That makes “getting it done” more valuable than “getting a better recommendation.”

A small number of consumer-life ecosystems (notably Alibaba and Meituan) span multiple verticals, so an agent can complete more end-to-end workflows inside a single service universe. ByteDance’s OS-level direction is different, and that is precisely why it runs into a different class of constraints when it tries to operate across apps that don’t share incentives. Within a single ecosystem, closed integration makes experimentation faster because payment, identity, fulfillment, and service rules can be coordinated end to end. Chinese consumers have repeatedly embraced new interaction patterns, from QR payments to super-app services, lowering the behavioral friction of delegation. The recent
Please, Log in or Register to view URLs content!
shows that 83% of respondents in China (compared to 39% in the United States) see AI-powered products and services as offering more benefits than drawbacks.

In many domains, new models are allowed to emerge first, with rules catching up once risks and patterns are clearer. That sequencing accelerates early trials relative to environments where uncertainty blocks deployment upfront. For instance, take
Please, Log in or Register to view URLs content!
: China released draft measures in spring 2023 and finalized interim rules in July 2023, effective in August, after the rapid emergence of gen AI companies and models and visible debate over governance. These conditions are hard to replicate in full. But the structural insight travels: agentic commerce scales where infrastructure, execution capacity, ecosystem coordination, consumer readiness, and governance are aligned. Companies headquartered in other parts of the world are making progress in this area: in early 2026, for example,
Please, Log in or Register to view URLs content!
to integrate Walmart’s inventory directly into Gemini, allowing for “agent-led” shopping journeys. The difference is that China’s tightly integrated “super-apps” and payment infrastructure provide the “plumbing” for these agents to scale much faster.

Please, Log in or Register to view URLs content!
 

mossen

Senior Member
Registered Member
I hope this awareness also exists at the highest levels. It would be quite disastrous if it were limited to lower ranks outside of Zhongnanhai

What the Iran war showed was basically "The Revenge of the Physical World". You could say the same thing with China's rare earth minerals ban. If there would be a massive cyberattack by the US against China then China could just cripple US industry overnight via the rare earth ban. Software cannot create molecules out of thin air and you certainly cannot print them via the central bank. And that's just one of many ways China can retaliate.

Besides, you have to ask yourself if you think that LLMs are a pathway towards true AGI. I don't think they are. And if they are not, then China being 6-12 months behind the US leading labs is hardly a disaster because there will necessarily be a ceiling some years down the road, unless we get a completely new paradigm, radically different from transformer architectures. But then everyone has to start over again.

There are too many terminally online people on Twitter who think the entire world economy will hinge on who has the best benchmark score for their LLM model. I hope people don't start to believe them on this thread. Let's touch some grass, folks.
 
Top