Algorithm system

Looking for partners2Lin Yuan even specifically sought out and watched the video of Dudu’s big boss discussing OpenAI.

Their big boss had returned from Silicon Valley and was considered a technical elite among the returning overseas Chinese at the time, having started his career as an engineer. However, technology is like martial arts; if you don’t practice it often, you’ll get rusty.

Therefore, that guy uttered an incredibly brainless statement: it’s impossible for Dragon Country to produce a local large language model that can rival OpenAI.

Lin Yuan scoffed at this.

However, given Dudu’s consistent behavior, such remarks from them were normal, as Dudu had never fought a tough battle.

If it weren’t for Google’s self-destructive actions, and if it weren’t for domestic protection, if there were fair competition, Lin Yuan could only think of a “hehe” regarding Dudu’s definite ranking strategy.

How can you expect a company without the gene for fighting tough battles to not have weak knees?

However, such views on AI large language models are not uncommon internationally, because OpenAI invested 20 billion, which intimidated later entrants.

From the perspective of later entrants, it’s almost impossible to compete with OpenAI, because AI requires continuous investment, and that 20 billion was just an initial, phased investment. What’s more terrifying is that the AI field is a global supply and demand area, with top AI talent flowing globally, making it impossible to reduce R&D investment through exchange rate differences.

This is similar to electronic components and energy sources like oil, where bulk transaction prices are common globally.

If you pay high-end AI talent in your local currency, they will turn around and go to Eagleland; isn’t earning US dollars more appealing to them?

Of course, these are past experiences. Lin Yuan does not deny that these experiences are wrong.

However, the AI field is far from mature; in other words, the AI field is still in a volatile development period.

This thing is not like the internal combustion engine, which has been played with for over a hundred years, with all technical paths already thoroughly understood. In fields where technical paths are already incredibly clear, the large amount of R&D costs sunk in the early stages will block the path of latecomers.

This is not only dependent on customer recognition in the market, nor solely on product maturity; most importantly, various patent barriers have already been set up along all technical paths, making latecomers stumble. Then, latecomers naturally have no way out.

Therefore, before the rise of New Energy vehicles, domestic cars were unable to enter the mid-to-high-end market, and even struggled above 150,000 yuan.

However, the AI field is a completely uncharted wilderness with no established path.

OpenAI seemingly invested 20 billion, but it did not create a feasible breakthrough technical path. Essentially, it is still rushing along the traditional AI training path, relying simply on the brute force brought by computational power.

Thus, with his sensitivity to technology, even if Lin Yuan didn’t have a system, he would still come to one conclusion: in a field that is a complete wilderness like this, black swan events are most likely to occur.

It’s like the Age of Discovery; in the vast, foggy ocean, who knows when an island might faintly emerge.

Comparing this to the technical realization of AI, what if someone comes up with a new algorithm or a new idea? After all, AI training is extremely demanding on the performance of algorithms and models.

Regarding the effort corresponding to computing power, anyone who has studied knows that studying hard does not necessarily guarantee good grades.

Therefore, Lin Yuan even felt that OpenAI’s strategy of aggressively stacking computing power might be The Other Shore’s open conspiracy. After all, The Other Shore can leverage their own advantages to the fullest in this way.

The world’s best GPU computing card companies are on The Other Shore, and the largest capital is also on The Other Shore.

Following a technology + capital-intensive route, there is indeed no place in the world that can outperform The Other Shore.

Lin Yuan plans to cooperate with a company that has integrity and dreams. Companies like Dudu, which have already knelt, are simply beyond help.

Soon, Lin Yuan contacted Chrysanthemum Factory.

He had high hopes for Chrysanthemum Factory because it is an ambitious enterprise.

But he soon discovered that having too much integrity and ambition was not good either.

Because Chrysanthemum Factory solemnly declared that they have their own pace.

Chrysanthemum Factory has its own understanding of AI large language models. Chrysanthemum Factory does not approve of pouring money into general large language models, but rather focuses on building an AI ecosystem around its own products.

Simply put, Chrysanthemum Factory’s current business is very broad, ranging from telecommunications equipment to cloud computing, from smart devices to car manufacturing, and even to lower-level operating systems and chip design. Chrysanthemum Factory is more concerned with the construction of the entire ecosystem.

As for its attitude towards AI large language models, it is to closely follow industry developments and then try to make large language models serve its own products. For example: using them in their own cars.

Chrysanthemum Factory is not interested in making big news with large language models. Their view is similar to Lin Yuan’s: there is still too much uncertainty in the AI field. And when facing these uncertainties, the best approach is to be down-to-earth.

Being down-to-earth means building AI around applications.

Lin Yuan strongly agreed with this. If he didn’t have a computing power system, he would certainly do the same.

After all, ordinary companies cannot make storytelling their core business. For an enterprise like Chrysanthemum Factory, which has suffered suppression and sanctions, being down-to-earth is crucial to their survival. Limited R&D funds cannot be invested in areas where monetization is not visible in the short to medium term.

But Lin Yuan also considered another layer: Chrysanthemum Factory has already been tormented by The Other Shore in the mobile phone, chip, and communication fields. If it were to achieve any major breakthroughs in the AI field, it would truly be attracting even more hostility. At that time, Eagleland would likely exert its utmost effort to target Chrysanthemum Factory.

Therefore, Chrysanthemum Factory, for strategic reasons, will certainly focus its efforts on breaking through the current critical area: the chip industry chain.

Moreover, their car manufacturing division is just beginning to flourish.

With these two considerations, they naturally had no interest in AI large language models.

However, Chrysanthemum Factory was willing to offer Lin Yuan an honorary technical advisor title. The other party specifically emphasized that it was honorary in nature, not an employment relationship.

This was naturally not because Chrysanthemum Factory was stingy, but out of politeness and courtesy.

Given Lin Yuan’s current reputation, Chrysanthemum Factory felt they didn’t have a suitable position for him. But since Lin Yuan had come to them to discuss cooperation, out of consideration for appearances, they at least showed friendliness to Lin Yuan.

Lin Yuan naturally declined the honorary technical advisor title. He respected Chrysanthemum Factory’s decision and had no intention of further communication.

So, another one was crossed off his list.

There were only a few domestic companies of a certain scale and strength left.

Coarse Grain Factory? Lin Yuan did consider it.

But he didn’t have much confidence in Coarse Grain Factory because its supply chain was not entirely in its own hands.

As the saying goes: when Snapdragon releases a generation, Coarse Grain Factory releases a generation.

The supply of core components is controlled by companies on The Other Shore, not to mention the messy equity structure behind Coarse Grain Factory. No matter how capable Lei Zi is, his arm ultimately cannot twist a thigh. Lin Yuan is now hated by The Other Shore’s capitalists, and he certainly wouldn’t walk right into their trap.

It’s very likely that the investors behind OpenAI, Tesla, and Coarse Grain Factory overlap. What would it mean if Lin Yuan went to Coarse Grain Factory? It would be like going to his enemies and saying, ”I want to make big news to mess with you.”

Lin Yuan certainly wasn’t that foolish.