Artificial Intelligence thread
- Thread starter 9dashline
- Start date
For me it would be cleaning the bathroom...
Simply, the Chinese haven't come up with any new/better AI product that's global and mainstream; even as many Chinese companies/institutions keep making claims of making LLM with more variables that's better and more powerful than others. Yet, no Chinese companies/person seem to be smart or innovate enough to push the forefront of AI to make revolutionary globally-reaching products. So why would any global medias reports on China's me2 but inferior products like Ernie and soon whatever name for their me2 but inferior product that is like SORA.
Not that simple. Maybe some are afraid they become a target like Huawei vilified and villinized. Huawei survive because it has deep pockets and the help of the government. How would a new enterprising Chinese A.I. developer going to make the product GLOBAL MAINSTREAM when it's going to be deemed as NATIONAL SECURITY RISKS before it can even take off or adopted by other countries in the world?
Let's not look at the business environment as if it's free from geopolitical issues and machinations that's been the norm in today's world.
Which essentially forces those ambitious Chinese A.I. developers to either sell their interest outside China or go establish their dreams outside China which means going to countries that are actively trying to butcher your own country metaphorically and literally. At the end of the day, you'll have to kowtow to Western Capitalists if they want to see their products mainstream.
Last edited:
It is that simple. Stop being pussies and release a product in China exclusively for China if you actually have something better than anyone else. Make them beg for access. They can’t because they don’t.
That could be the case as well. But maybe it's a combo of both? Rusk taking goes hand in hand with innovation. But, the fear of failure, embarrassed is still pervasive in my opinion which has a debilitating factor for China to come out with a product that's standard settings rather than the usual of playing catch up.
I mean, how do we explain that many of the people that work in A.I. LLLM in the U.S. are of Chinese origin and are Chinese? The air and water in America makes you more innovative? Lol
It isn't that simple.
Just look through this thread a bit (last 10 pages) and there are various example of Chinese AI products that doesn't really have a matching western one, while being a useful product that can boost productivity.
And yes, they aren't anywhere near as headline grabbing as something like SORA, but still very impactful (if not more so).
Not to mention chatgpt, SORA etc. is also just like some 6ish months ahead a Chinese equivalent, and it's very unlikely that stuff by OpenAI will really have such an impact or importance of being 6 moths earlier (and no, I don't think the current LLM is anywhere close to something like generative AI, it might become part of the puzzle, but it would just be that, a puzzle part, one in many needed).
The world is a big place. There are plenty of (non-Western) places where a Chinese AI LLM product could thrive.
In the open source community, Yi and Qwen have been quite good. If their next product retains the same increase in capabilities there is no reason why they couldnt be commercialised.
Its just a matter of value for money
No, Chinese firms simply underinvest in pure NN research, they mostly focus on applied stuff and short-term results. For example, Microsoft Research Asia is based in Beijing yet produced many ground-breaking works - ResNet (most cited DL paper), Swin Transformers, etc. They also created LightGBM which is one of the most popular gradient boosting libraries now. So yeah, this shows that with enough investment Chinese researchers can and will create strong research papers.