And then the AI bubble burst. Or, as one meme put it: “ChatGPT just lost its job to AI.” Instead of obsessing over TikTok’s 24-hour banning in the United States, the world is in thrall of another Chinese app, DeepSeek. This artificial intelligence (AI) startup recently released its latest chatbot model with many capabilities that match OpenAI’s ChatGPT – except developed at a fraction of the cost.
Instead of the hundreds of millions of dollars that US firms have invested in training advanced AI models, DeepSeek says it only spent $5.6 million developing its R1 model. This is estimated at 3% to 5% of what ChatGPT costs to develop.
The bubble didn’t just burst, it blew up. Since ChatGPT arrived on the scene on the last day of November 2022, there has been an arm’s race by US firms to build bigger data centres to run the bigger large-language models (LLMs) that need more and more computing power. Or, as the industry itself calls it in flagrant disregard of grammar, “compute”.
To say Nasdaq fell off a cliff is an understatement, as was the speed of the fall from grace of the tech stocks primed to take advantage of this artificial intelligence hype cycle. On Monday, the Nasdaq saw $1-trillion, or 3.1%, wiped off it. Chipmaker Nvidia, which briefly eclipsed Apple and Microsoft last year to become the most valuable listed company, lost $600 billion alone. This is a dubious new record in its own right, the biggest one-day loss in US history. Ouch.
Google’s owner Alphabet lost $100 billion, while Microsoft, which owns a significant stake in OpenAI, lost $7 billion. SoftBank – which is part of US President Donald Trump’s big announcement about building AI data centre capacity with OpenAI and Oracle called Stargate – lost 8% on Monday and 5% on Tuesday. More ouch.
Silicon Valley venture capitalist Marc Andreessen, who founded the Netscape browser, called this AI’s “Sputnik moment” – when the Soviet Union beat the better-resourced US into space with its Sputnik satellite.
Indeed, it certainly seems like the AI bubble burst. The sky-high projections for tech firms have been undermined by a little Chinese quantitative analysis firm, called High-Flyer, which decided to build its own chatbot. Instead of creating a large-language model, which is a massively expensive undertaking, DeepSeek used the output of another LLM to build its R1 model, a process known as distillation.
DeepSeek’s release was akin to that of OpenAI’s ChatGPT two years ago, which, as DeepSeek has now done, also became the most downloaded app on Apple’s app store.
DeepSeek’s release proved that it’s possible to create advanced so-called chatbot models using a fraction of the resources that the big US AI firms have spent. Companies like OpenAI, Google, Microsoft, and Facebook have invested tens of billions of dollars in high-end servers and powerful chips like Nvidia’s H100, which is specifically designed for AI use.
But DeepSeek, which published a paper on how it created R1, says it used Nvidia’s much less powerful H800 processor, which the chipmaker created because of restrictions on what US technology could be exported to China. These H800 chips were also banned last October.
“I don’t believe it’s $6 million, but even if it’s $60 million, it’s a game changer,” Umesh Padval, MD of Thomvest Ventures, which invests in AI firms told Wired. “It will put pressure on the profitability of companies which are focused on consumer AI.”
As if to confirm this, this week DeepSeek released its own image-generating service called Janus-Pro, which already matches the output of OpenAI’s market-leading DALL-E 3.
Last year Goldman Sachs cautioned whether the estimated $1 trillion in capex for “data centres, chips, other AI infrastructure, and the power grid,” which already has “little to show for it so far,” and might not “ever pay off.”
OpenAI CEO Sam Altman was full of praise. “DeepSeek’s R1 is an impressive model, particularly around what they’re able to deliver for the price,” he tweeted. “We will obviously deliver much better models and also it’s legit invigorating to have a new competitor. We will pull up some releases.”
Last month, Altman stood next to Trump as he announced Stargate, a project to build AI infrastructure (which includes OpenAI, Oracle and SoftBank), that will invest $500 billion. Trump recently called DeepSeek a “wake-up call” and pointed out that it is “AI from a Chinese company”. Talk about focussing on the wrong Chinese app. TikTok may be silently relieved there is a new ‘baddie’ in town.
“The race is on,” says Bank Zero chairman Michael Jordaan, himself a tech-savvy startup investor since he resigned as FNB CEO a decade ago. “What DeepSeek demonstrated is that AI models can be built with far lower training costs on much less advanced and expensive chips. This also requires far less energy,” he tells Business Day.
As he quite rightly points out, “the DeepSeek method is now open-sourced, so it is no longer just Chinese. Everyone can use their insights for free if they want to. So, the US AI industry will catch up fast.”
However, the most cogent thing said about AI was by Scott Hanselman, Microsoft’s vice president of its developer community, at an AI event in Sandton last month. “We have to push back on the idea that AI is smarter than us. It’s artificial but it’s not real intelligence. We are the real intelligence,” he said, before adding: “Is the spanner the thing we hired or is it the plumber that we hired?”
This article first appeared in Business Day
