The launch of ChatGPT in late 2022 generated a boom in the technology world. What do you think about it?
Just within two months, ChatGPT had 100 million users, the fastest growing rate in the history of technological products. It took Facebook four years to reach that goal.
I think the breakthrough was made thanks to the LLM (large language model). It is excellent, if compared with previous content processing products.
LLMm has created a new UX/UI (user experience/user interface). As Microsoft pointed out, there have been three user interface revolutions. The first revolution occurred when users approached information with keyboard and mouse (PC). The second occurred when touch screens appeared (smartphone). And now, chat.
This new interface is likely to completely change the way users interact with content, which is very important for technology companies.
How have Vietnamese reacted to AI Chat products?
ChatGPT still doesn’t have an official operation in Vietnam (users cannot create accounts and make payment in Vietnam), which means they are not focusing on Vietnam. However, we have noticed that there is a high number of people using ChatGPT in Vietnam via Coc Coc.
We find the market relatively new and we believe that there are many opportunities so we have decided to join it, with two orientations. First, focusing on the Vietnamese market. Second, integrating with a search engine.
So, will Coc Coc have a different development orientation from rivals when deploying its products?
If you make a comparison, you may see that Coc Coc’s product is closer to Microsoft’s Bing AI, because Microsoft also uses LLM, integrated with a search engine.
I guess that chatbots and search engines may merge one day in the future. At first, summarizing the information of one page, later summarizing the info of many pages, and after that, the chatbot responds to users’ wishes much more quickly than search engines.
At present, we can see that Bing AI is integrating the two features. Meanwhile, we are deliberately separating the features into Coc Coc AI Chat and Coc Coc AI Search.
If you try Bing’s AI chatbot, you find limitations in the production of highly creative content, because of the mechanism which always relies on a search engine. However, generative AI is interesting and it sometimes goes crazy. One chatbot, if only relying on a search engine, would be weak at creativity.
What is the difference in deploying the new product and previous products?
The products we deployed previously, search engine or browser, were not new, so we could clearly imagine users and business models and we knew how we needed to compete. But with AI chatbots, things are completely new, and many users still don’t know how to make the best use of it.
So, what do you think could be the feasible business models for AI chatbot based on LLM?
I believe that even the big players are still not sure about that. As there is still no clear business model, the revenue prospects are not good. The investment in this field, for some big players such as Microsoft, is aimed to defeat Google’s business model.
As I said above, LLM creates a new user interface which may replace the current major tools of the internet, the browser and search engine, because these are the two basic tools to access content.
Some analysts believe that Google still doesn’t make this product because it may destroy exactly its current business model. Search engines are now the major money-making machine.
The scenario most likely to happen, I think, is that there would be only several companies maintaining the LLM model in the world, after that they go selling the basic service – LLM. Meanwhile, using LLM to create products with their own business models is something that many companies can create.
What is Coc Coc’s approach when there is still no clear business model?
Our approach is to launch the product first. The first goal is to learn how to apply this new technology. More importantly, Coc Coc wants to learn about users. When users try the product, Coc Coc will know how they use it, and how the product brings value to them. This will help find use cases.
Trong Dat