Call it the end of the beginning of the AI boom.
Since mid-March, financial pressure on several AI startups has had a negative impact. Inflection AI, which raised $1.5 billion but earned next to nothing, has abandoned its original business. Stability AI has laid off employees and parted ways with its CEO. And Anthropic has raced to close the roughly $1.8 billion gap between its modest sales and huge expenses.
The artificial intelligence revolution, it is becoming clear in Silicon Valley, will come at a very high price. And the tech companies that have bet their futures on it are scrambling to figure out how to close the gap between those expenses and the profits they hope to make in the future.
This problem is particularly acute for a group of high-profile startups that have raised tens of billions of dollars to develop generative artificial intelligence, the technology behind chatbots like ChatGPT. Some of them are already realizing that competing directly with giants like Google, Microsoft and Meta will require billions of dollars – and even that may not be enough.
“You can already see the writing on the wall,” said Ali Ghodsi, CEO of Databricks, a data warehouse and analytics company that works with artificial intelligence startups. “It doesn't matter how great what you do is: is it commercially viable?”
While a lot of money has been burned in other tech booms, spending on building AI systems has shocked tech industry veterans. Unlike the iPhone, which kicked off the latest technology transition and cost a few hundred million dollars to develop because it relied largely on existing components, the creation and maintenance of AI models generative costs billions. The cutting-edge chips they need are expensive and in short supply. And each query from an AI system costs much more than a simple Google search.
According to PitchBook, which tracks the industry, investors have poured $330 billion into about 26,000 artificial intelligence and machine learning startups over the past three years. This is two-thirds more than the amount spent to fund 20,350 AI companies from 2018 to 2020.
The challenges affecting many new AI companies are at odds with early business results from OpenAI, which is backed by $13 billion from Microsoft. The attention it generated with its ChatGPT system allowed the company to create a company that charges $20 a month for its premium chatbot and gave companies a way to build their own AI services with the technology that drives his chatbot, called large language. model. OpenAI reported revenue of about $1.6 billion over the past year, but it's unclear how much the company is spending, two people familiar with the company's business said.
OpenAI did not respond to requests for comment.
But OpenAI has also faced challenges in scaling up sales. Companies fear that AI systems could generate inaccurate responses. The technology has also been troubled by the question of whether the data that supported the models violated copyrights.
(The New York Times sued OpenAI and Microsoft in December for copyright infringement of news content related to artificial intelligence systems.)
Many investors point to Microsoft's rapid sales growth as evidence of artificial intelligence's commercial potential. In its most recent quarter, Microsoft reported an estimated $1 billion in sales from artificial intelligence services in cloud computing, up essentially from zero a year ago, said Brad Reback, an analyst at investment bank Stifel.
Meta, on the other hand, doesn't expect to make money on its AI products for years, even as it ramped up infrastructure spending to $10 billion this year. “We're investing to stay at the forefront of this space,” Mark Zuckerberg, Meta's chief executive, said on a call with analysts last week. “And we're doing this at a time when we're also scaling the product before it makes money.”
AI startups have been challenged by this gap between spending and sales. Anthropic, which has raised more than $7 billion with backing from Amazon and Google, spends about $2 billion a year but takes in only $150 million to $200 million in revenue, two people familiar with the matter said. company financial data, who requested anonymity because the data is private.
Like OpenAI, Anthropic has turned to partnerships with large, established technology companies. Its CEO, Dario Amodei, has been courting clients on Wall Street and recently announced that he is working with Accenture, the global consultancy, to create custom chatbots and artificial intelligence systems for businesses and government organizations.
Sally Aldous, a spokeswoman for Anthropic, said that thousands of businesses use the company's technology and that millions of consumers use its publicly available chatbot, Claude.
Stability AI, which generates images, announced last month that its founding CEO, Emad Mostaque, had resigned, just a week after the resignations of three researchers who were part of the five-person team that developed the company's original technology.
It was on track to generate about $60 million in sales this year against about $96 million in costs from its imaging system, which is available to customers from 2022, said a person familiar with familiar with his business.
Stability AI's financial position appears better than that of language model makers such as Anthropic because developing image generation systems is less expensive, AI investors said. But there is also less demand to pay for images, so sales prospects are more uncertain.
Stability AI operated without the support of a tech giant. After raising $101 million from venture capitalists in 2022, it needed more funding last fall but struggled to prove to investors that it could sell its technology to businesses, said two former employees, who declined to speak publicly because they did not they were authorized to do so. It raised $50 million from Intel late last year but still faced financial pressure, they said.
As the start-up grew, its sales strategy changed, these people said. At the same time, it was spending millions a month on IT costs. Some investors have been pressuring Mr. Mostaque to resign, according to one investor, who declined to speak publicly about a personnel matter. This month, after his resignation, Stability AI made layoffs and restructured its business to put the company on “a more sustainable path,” according to a company memo reviewed by The New York Times.
Stability AI declined to comment. Mr. Mostaque declined to discuss his exit.
Inflection AI, a chatbot startup founded by three AI veterans, has raised $1.5 billion from some of the biggest names in tech. But a year after introducing its AI personal assistant, it had almost no revenue, according to one investor. The Times reviewed a letter Inflection had sent to investors saying that further fundraising was “not the best use of our investors' money, especially in the context of the current frothy AI market.”
At the end of March it abandoned its original business and largely disappeared into Microsoft, the world's most valuable public company.
Microsoft also helped fund Inflection AI, whose CEO, Mustafa Suleyman, rose to fame as one of the founders of DeepMind, a foundational artificial intelligence lab that Google acquired in 2014. Suleyman founded Inflection AI together with Karén Simonyan, a key researcher at DeepMind. and Reid Hoffman, a leading Silicon Valley venture capitalist who helped found OpenAI and is on Microsoft's board of directors.
Microsoft and Inflection AI declined to comment.
The company was full of talented AI researchers who had worked at places like Google and OpenAI.
But nearly a year after launching its AI personal assistant, Inflection AI's revenue was, in the words of one investor, “de minimis.” Essentially nothing at all. It couldn't continue to improve its technologies and keep pace with Google and OpenAI's chatbots unless it continues to raise large sums of money.
Now Microsoft is swallowing up most of its staff, including Mr. Suleyman and Dr. Simonyan.
This is costing Microsoft more than $650 million. But unlike the Inflection AI, it can afford to play the game for the long term. It has announced plans to have staff build an artificial intelligence lab in London, working with the kind of systems that start-ups hope will break through.
Erin Griffith contributed to the reporting.