Startup AI is attracting massive stream of capital but also facing an alarming failure rate. The expert recommended that startup look for his own corner instead of throwing eggs.
Great Success but Failure Also Much
When artificial intelligence becomes a global technological hub, everything turns to them. That's why the series of startup AI was born and easily mobilized capital, in the context of investors still tied to many other fields.
In particular, in 2024, AI startup companies in the United States received 97 billion investments, taking nearly half of their total funding for American startup companies that year, according to PitchBook.
In addition, in the quarter of 2024, the global startup AI mobilized $11.8 billion, occupying 30% of the total venture capital in this quarter, according to Stocklytics.
However, unlike the cheap capital phase, startup just calls the capital live because it's the ace to media, recruit good people and campaign the next round. At the present time, capital alone is not enough. A report from CB Insights points to a disappointing reality: As far as 80% startup AI fails in the first 23 years.
The cause of this condition is partly due to intense competition. According to a census from TechCrunch, there are currently more than 1 million AI models worldwide from major companies in artificial intelligence such as Google, OpenAI to such startups as Anthropic, xAI.
But the key thing that prompted AI companies to quickly fall off their horses was that they were running towards a trend without clear strategy.
Many startup AI was born only to take on the technology wave, but lack the practicality or no sustainable competitive advantage, Mr. Hônzhen, technologist at TP.HCM, remarked.
Don't risk following the big boys
OpenAI's birth was like a steel punch into the global technology industry. It spurs the race from the majors in the technology village such as Amazon, Meta, Microsoft, Google... to the start-up companies into artificial intelligence.
However, in the White Book from the start-up firms to the Opening Races through the ecosystem of the Post-City Business School of Cheung Kong China, Professor Jon Koch, in charge of Management Practices, Operations, Change, and Industry at CMGSB and the former Director of Operations of JD.com, explaining, the use of AI's power at large is not always easy.
The environment today is a challenge for AI startup companies because first of all, the power of computing is extremely limited. They don't have the same computing power as digital giants like Amazon and Meta.
Similar to data. It's hard to imagine that the AI-based startup companies can compete with the giants who are responsible for the rich data network they have.
For example, an AI app for a small store. They need large amounts of data from within the ecosystem as price and supply; but they also need wider ecosystem data, such as customer demographic information, local shopping behavior and optimum store position, interpretation professor.
Koch predicted that this is the reason why the start-up companies based on AI are increasingly deep into two areas: the first is solutions dedicated to the industry, such as AI for pharmaceuticals, AI for marketing and AI for retail; and the second is AI for the compilation.
In Vietnam, running AI development was even more difficult in the lack of a million AI engineers. So, according to experts, the startup Vietnam should not venture in large models.
The AI force in Vietnam was not ripe enough to meet large-scale projects, Dr. Greentown Manor, instructor of Information and Media Technology School, the University of the Indigenous Sciences, remarked in the conference on AI recently.
Related Posts
