ChatGPT traffic peaked, AI revolution collapsed? Perhaps OpenAI is playing a big game

Original source: GEM Observation

Image source: Generated by Unbounded AI‌

This week, the A-share AI sector can be said to have fallen from the altar, and the sector showed a decline. During the period, the artificial intelligence index (931071) fell by 7.87%.

In terms of news, Kunlun Wanwei’s huge reduction of holdings dampened sentiment before, and then there was weak ChatGPT visit data—this is regarded as the main catalyst for the weakening of the sector.

At this moment, any turmoil may become an amplifier of panic, and doubts are endless-is this wave of AI market coming to an end?

To answer this question, we have to go back to the source of the above data.

According to data from the website data analysis tool imilarWeb, in January, the number of visits to the chatbot ChatGPT increased by 131.6% month-on-month, and in May, this figure changed to 2.8%. ** As of June 20, two-thirds of June has passed, and the number of visits is about 38% less than May. **

According to the data compiled by the consulting service agency AceCamp, almost all AI chatbots have stagnated in the growth of the number of users, and even began to show a downward trend.

However, some people are skeptical of the claim that ChatGPT traffic has peaked. The self-media "Little Bear Runs Fast" pointed out that the formula of multiplying the number of monthly active users by the average duration is unreasonable in itself. The traffic of ChatGPT has not experienced the "cliff-like decline" as described by some self-media. At present, it is generally stable. To B access is also in progress.

The introduction of "Little Bear Running Fast" on the Flush Investment Advisory platform is "the chief analyst of the computer industry of a certain brokerage".

Flow stagnation is not unexpected OpenAI may be playing a big game

Judging from OpenAI's recent statements and practices, it is not surprising that the company intends to "limit the flow" and the number of visits is stagnant. **

As early as April 5, OpenAI stopped the sale of ChatGPT Plus paid items. Then at the AI Congressional hearing on May 16, OpenAI CEO Sam Altman complained about "happiness troubles": **Due to the bottleneck of computing power, it would be better if fewer people used ChatGPT. **

What it said, OpenAI chose to take the initiative to "dissuade", **Recently, ChatGPT has been banned on a large scale again. **

A netizen received an email from OpenAI: Due to the suspicious behavior of the account, the Plus account was blocked to ensure the security of the platform

In the minutes of the closed-door meeting that flowed out in early June, **Altman further clarified OpenAI's product strategy-making ChatGPT a super-intelligent personal work assistant, attracting more companies to access its API instead of competing for customer traffic. **

**In order to fulfill the promise of "not competing with customers for traffic", OpenAI actively promotes the upgrade of API service fee reduction, **has recently announced a major update of its GPT model API, including new function call functions and reduced usage costs and many other things.

**Since OpenAI opened its plug-ins, the number of plug-ins has been increasing rapidly. **According to the statistics of foreign netizens, the total number of ChatGPT’s latest plug-ins has reached 589, an increase of more than 6 times in one and a half months.

All in all, **For OpenAI, building an AI ecosystem is far more important than controlling C-side traffic. **

The world beyond ChatGPT is vast: the application of a hundred flowers blossoms and payment is gradually becoming a trend

Standing alone is not spring, and this is true for any industry.

Large model-related applications are not limited to chatbots. At present, there are AI applications or AI technology access in the fields of games, media, embodied robots, office, medicine, education, e-commerce, and financial technology:

Tesla, Google, and Microsoft are all trying to inject soul into robots with large models;

Discord, Snap, BuzzFeed and other media companies develop their own AI products through OpenAI;

Ubisoft, Epic, Tencent, Microsoft, Roblox and other big companies are competing to show their AIGC technology in the game field;

Overseas leading manufacturers such as Google and Microsoft deeply integrate large-scale models with their own office products;

AI medical care has spawned a number of well-known companies such as Nuance and IBM Watson, and iFLYTEK, Baidu, etc. are also laying out large vertical models in the medical field...

However, the increase in users that AI brings to companies and individuals, and the implementation of the corresponding payment model, really let the industry see the hope of commercialization, such as:

The weekly visits of the data service software lamaindex.ai increased by 122.69% year-on-year;

After the overseas office software Notion introduced the AI function, the ARPU value increased by 50-100%;

A female Internet celebrity in the United States launches a digital human version of herself and charges by the minute. It is estimated that her annual income will increase by 60 times.

**In less than a year since the AIGC wave started, the commercialization trend in the AIGC field has become extremely obvious. **Current commercial AIGC tools such as Notion AI, Adobe Firefly, Midjourney, etc. all use subscription as the main payment model.

Among them, Adobe, a veteran PC software company, announced its second-quarter results in June, showing that the company's revenue hit a record high in the three months ended June 2.

It is worth noting that with the accelerated arrival of the AIGC payment era, not only product developers have benefited from it, but data providers have also started charging ideas. Twitter and Reddit, the "US version of Post Bar" have set higher API charging standards.

The wheels of the technological revolution are rolling forward, don't "get up early, catch up late"

Funds and emotions in the secondary market often come and go quickly, and the movement of the primary market can better reflect the true temperature of the industry.

In China, many bigwigs including Zhou Bowen, the former AI leader of JD.com, Li Yan, the core figure of Kuaishou AI, and Li Kaifu, chairman and CEO of Sinovation Ventures, formed a group to start a business.

Overseas, many star projects such as Cohere, Runway, and Replit have emerged. According to data from CB Insights, an American research institute, 13 AIGC companies have become unicorns (startups with a valuation of more than US$1 billion), and the average time it takes them to go from starting a business to becoming a unicorn has been cut in half. **

Looking at home and abroad, the global development of AI has been regarded as a strategic policy by many countries and regions. **

Domestically, the Artificial Intelligence Law has been included in the legislative work plan;

The Biden administration has announced a series of new initiatives around the use and development of artificial intelligence in the United States in May;

The Japanese Ministry of Internal Affairs and Communications and the Ministry of Economy, Trade and Industry will merge AI-related guidelines, or mention generative AI for the first time;

The plenary session of the European Parliament has voted to approve the draft mandate for the Artificial Intelligence Act.

It is true that ChatGPT is the most popular AIGC application at the moment, but if we focus too much on a single product, it is easy to blind our eyes and ignore the vigorous vitality of the entire AI industry, from application to policy and investment.

In other words, **Even if ChatGPT visits enter a bottleneck period, it is too early to talk about the end of the AI revolution at this time. **Although OpenAI's traffic has peaked, the company pays more attention to ecological building and improving profitability, rather than illusory visitor data.

At a recent book reading session, Microsoft (China) Chief Technology Officer (CTO) Wei Qing said, "I believe many people have tried ChatGPT and put it aside, because it is basically related to our work. The Internet is still fragmented. But I still hope that everyone will not 'get up early and catch the late episode', because this is a paradigm revolution that will bring subversive changes.”

Yang Jie, chairman of China Mobile, has the same opinion - AIGC may trigger a new round of technological revolution, and this is just the beginning.