Thanks to the appearance of cloud computing and distributed digital infrastructure, one micropoene prize is far from a novel concept. Inexpensive on -demand calculations, remote collaboration, payment processing APIs, social media, and e -commerce markets are all easy to go as an entrepreneur.
But how about expanding that one business to something with a lot of flesh?
Historically, this was an immeasurable tough work for the skills and resources needed not only to expand the product, but also to grow and maintain a rich customer base. However, AI agents can solve the world solo processing.
AI agents are to embed human workflows in software and release more things in less humans. You can assign tasks to agents and make decisions with various degree of autonomy. Multiple AI agents can even cooperate with complementary tasks and open the actual work completely autonomously.
In an interview with Alexis Ohanian, a co -founder of Reddit last year, Openai’s Sam Altman predicted this accurate scenario.
“In my small group chat high -tech CEO friend, there is a pool for this bet in the first year that there is one company,” Altman says. “This would have been unimaginable without AI, and it happens now.”
In the discussion at the Annual Meeting of the World Economic Forum held last week in Davos, entrepreneurs and investors also discussed the outlook for a single $ 1 billion company.
We trust
The recent history has revealed the majority of hundreds of dollars of large -scale companies. Microsoft paid $ 2.5 billion for the reported Minecraft Maker Mojang reported. Facebook acquired Whatsapp for $ 19 billion if the messaging app maker had only 55 employees. Two years ago, Facebook purchased Instagram for $ 1 billion, and was only 13 employees.
This proves that Internet technology has already created a huge company with the minimum employment. But this is not the same as a single unicorn.
Kanjun QiU, a CEO of AI Research Lab IMBUE, is an agent that can be inferred and coding, so the most likely AI is the most likely business is a self -service. I think there is.
“I think it’s the easiest, and the first thing that doesn’t require a team that is on a large -scale market (either consumer or promot),” was in a panel discussion. I explained. “I think that going to the market is actually one of the difficulties to automate all of these relationships with others.”
Regarding sales generation, winning is not always a good product. The people behind the product did a good job rather than building trust with customers. Therefore, even if you need to actively sell products, you may need to upgrade the staff.
“I think the trust from humans to humans is still very needed and very important,” QiU added.
Benjamine LIU, a CEO of AI Drug Development Company Formation Bio, is bullish on the growth of AI and since then.
“I think we live in one of the most exciting fields, the building company,” LIU said in Davos. “We have PHD-level intelligence in pockets. We are starting to see the AI system working on the whole team. In that world, AI-Native companies have a considerable advantage. I think.
However, LIU repeated QiU’s emotions. The possibility of one mega business is real, but in fact it may not be meaningful from a business or entrepreneurian point of view.
“The possibility of getting there is faster than people think,” LI said. “In my opinion, being an entrepreneur is a kind of lonely journey, and it takes a long time because I want a co -founder. Companies are still started by humans. For some people. I want you to share your trip. “
So, in reality, we may reach the place where companies have always started, that is, the established team with a complementary skill set. However, instead of expanding gradually, they maintain the AI agent plugging in the gap.
However, even if the legendary first -person unicorn does not happen, there is almost no question that approaching Agent AI freight trains will greatly confuse labor.
“AI era of AI employees”
If all this looks hypothetical, think again. Agent AI has already participated in the workforce in the form of a software engineer such as Harvey and Cognition’s Devin.
AI’s sales agents are also booming, and as shown by the Dystopian Billboard AD of San Francisco, companies supported by craftsmen and other VCs are proud to replace human labor.
Many other companies also have a foundation for agent AI to prosper.
Lattice is the last of $ 3 billion of HR and human resource management platforms, and is further advanced by providing official digital workers. The manager was assigned.
Sarah Franklin, who joined Lattice as a CEO last year, called this transition “great collaboration” and worked side by side with human and AI agents. What this means is to manage these agents in the same way as humans to promote transparency and accountability.
“We want to give priority to the success of people as a primary. It is important to understand what they are doing when cooperating with the AI agent,” Franklin explains in Davos. did. “I don’t mean that AI is a human. It is necessary to clearly identify where AI is. AI speaks on behalf of brands and people, and make decisions on behalf of brands and people. You need to track it to integrate with other systems.
But what does this mean to society if a company can operate on a large scale without important human labor? People need to make money. They need their goals -If people can’t work, will society break their own?
Like the previous Industrial Revolution, a general copy of the AI Revolution is that new employment occurs in the long term.
“There are many employment creations,” said Mitchell Green, the founder of the investment company lead edge capital. “Considering when the iPhone was released in 2007, Uber and Airbnb are currently a $ 100 billion company. (They) would not have existed before this. We can have the opportunity yet. They will be the next huge business.
But that doesn’t mean that there is no pain in the short term in the short term. Also, as already seen in China’s AI -like DEEPSEEK, the speed of AI forward is important in terms of cost -to -performance ratio of AI models. And this can be an important differentiation factor compared to the previous industrial and technology revolution. We may not be able to adapt quickly enough.
“I think there are a lot to talk about re -training and up skill,” Li said. “However, there is a very unique thing that the development pace and how fast these models are, especially the work of these AI systems working on the whole team.”
“AIS manager”
Between the panels, there was a common consensus that not only live with AI, but also to learn how to use it in the workplace, regardless of the rate of change.
For example, on You.com, companies can access APIs and use real -time web searches to large language models (LLMs). We provide our own agent suite to specific tasks, but companies may create their own custom agents, select a priority AI model, and provide instructions based on data sources that need to be connected. can.
“We will be the first generation to manage AI with people as a CEO,” said Davos. “But the most interesting change here is that all the individual contributors, all employees, are all employees of AIS, and in that sense, everyone is a kind of entrepreneur. It will be.
Therefore, JU is still doing a lot of things about seeing a true unicorn company. However, the principle behind emotions has already been proven, as seen at a very valuable WhatsApp, which worked for each employee for $ 345 million when Facebook was purchased.
Even NVIDIA, which has a market capitalization of over $ 3 trillion, has a relatively thin labor for less than 30,000 employees. This is equivalent to about $ 100 million per employee.
It is difficult to see AI heading north, north, north, south to south with appropriate type of company and appropriate execution. But perhaps it depends on whether one person wants to build alone.
However, it is completely different questions whether society is ready to process it.
