ChatGPT visits decline for the first time, generative AI is going to cool down?

Article source: Sanyi Life

Although ChatGPT has cooled down at the consumption level, it does not seem to change the optimistic attitude of enterprises towards generative AI.

"The popularity of ChatGPT is declining", this is not nonsense, but the data released by Similarweb, a professional network analysis company. According to their recent data, in the first five months of 2023, ChatGPT’s global visits increased by 131.6%, 62.5%, 55.8%, 12.6%, and 2.8% month-on-month, and by June, ChatGPT’s The number of visits fell by 9.7% month-on-month, which was also the first decline since its launch last fall.

The data from Similarweb also reflects the popularity of ChatGPT on the Internet. As the fastest growing consumer application in history, only two months after the launch of ChatGPT, its monthly active users have exceeded 100 million at the end of January this year. At the beginning of this year, ChatGPT was widely discussed among netizens.

Although ChatGPT has cooled down at the consumer level, it does not seem to have changed the company's optimistic attitude towards generative AI.

At the same time, IT solutions integrator Insight Enterprises and research firm The Harris Poll jointly conducted a survey of 405 full-time executive-level employees aged 25 and over in the United States at companies with more than 1,000 employees. Respondents in job titles or higher, conducted a targeted survey of corporate attitudes toward generative AI. The results showed that 72 percent of business leaders from Fortune 500 companies plan to adopt generative AI within the next three years to improve employee productivity.

Not only the leaders of overseas companies are optimistic about generative AI, but also in China. A few days ago, NetEase CEO Ding Lei stated at the Zhejiang Provincial Platform Economy High-Quality Development Conference that exploring the innovation and application of AI large models at the fastest speed is NetEase’s top priority. The package platform will also provide AI new jobs for 100,000 people this year. Behind such disparate data, is the artificial intelligence bubble coming again, or is generative AI promising in the future?

The best description of artificial intelligence today should be "the future is bright, but the road is tortuous". Now that consumers are not interested in ChatGPT, Similarweb is not the first research organization to discover that it is even a public fact that ChatGPT is not popular. The reason why consumers lose interest in ChatGPT is also very simple. At this stage, it is indeed a bit exaggerated, and its role has been exaggerated by the outside world.

In fact, ChatGPT is not strong artificial intelligence (StrongAI), nor is it general artificial intelligence (AGI), but a large language model (LLM), a natural language processing model based on artificial intelligence technology, just because the parameter scale of ChatGPT is very It is so big that it has the phenomenon of intelligent emergence, which can process large-scale natural language texts, has the ability to generate natural language, and has obtained a certain reasoning ability. On the whole, ChatGPT is actually just the result of Li Dazhuanfei, rather than a new breakthrough in the basic level of AI.

In essence, ChatGPT pushes deep learning to the extreme in large language models, but this is only the extreme of deep learning algorithms, not the extreme of artificial intelligence. ChatGPT and other LLMs can only generalize large-scale data and deduce some common problems, but as long as the content entered by the user is relatively unfamiliar in the database, they can only start to talk nonsense.

The limitations of ChatGPT have led to real practical applications. It is not as easy to use as everyone thinks. Even if it is the best dialogue robot ever, it is nothing more than that. ChatGPT can indeed write copywriting, but do you dare to use the results directly in commercial scenarios? Obviously not, so you will still check the content output by ChatGPT to ensure that copyright risks and legal compliance are excluded. Similarly, ChatGPT can also compile code, but programmers dare not directly deploy the code it generates.

For example, it is reported that in March this year, many departments within NetEase, including NetEase Games, NetEase Cloud Music, NetEase Youdao, etc., began to try AI to improve efficiency, connected to the ChatGPT API, and used ChatGPT for testing, but the result was It's a mixed blessing. In fact, due to the lack of reasoning ability, ChatGPT's content production ability in the face of complex scenes is greatly restricted. And the most deadly thing is that because some users use ChatGPT's networking function to bypass the paywalls of other websites, OpenAI has urgently removed this function, further limiting ChatGPT's ability to obtain information.

So in the final analysis, there are actually very limited scenarios where ChatGPT’s strength can be used now. When the novelty of users passes, they will naturally not be interested in it. The reason why companies value ChatGPT or generative AI is also very simple. Even though they still have great limitations at this stage, replacing labor with machines is undoubtedly the most perfect solution for cost reduction and efficiency increase strategies. Therefore, when users value the current experience, while enterprises focus on the future, this is the reason why generative AI is now in the middle of ice and fire.