“The bursting of a generative artificial intelligence bubble is not on the agenda”

LGenerative artificial intelligence (GAI) like ChatGPT is generating unprecedented enthusiasm, amplified by the good results of Big Tech, especially Microsoft, Amazon, Google and Facebook. As 2024 begins, Nvidia, the leading maker of AI chips, is posting spectacular results, with revenue up 265% compared to 2023, propelling its valuation to more than $2 trillion, surpassing the oil company Aramco, and reaching third place in the world in terms of capitalization.

Some see AGI as the new oil. Skeptics compare this surge to the dot-com bubble at the turn of the millennium and the telecommunications bubble of the 1990s, both of which burst. Between January 1996 and March 2000, the price of Internet stocks had increased by 1,000%!

In 2024, massive investments in AGI and outsized forecasts could indicate overheating, particularly as the top ten companies in the S&P 500 are more overvalued than those in the 1990s tech bubble. Early speculation of the Internet had led to the creation, by new players, of very expensive oversized network infrastructures.

A significant operating cost

Today, IAG’s infrastructure is already in the hands of well-established Big Tech. And the current shortage of IAG chips reduces the risk of inflation in the construction by adventurers of excessive new infrastructures. As for current funding for IAG, it mainly comes from Big Tech and a few large venture capital firms.

Read the column | Article reserved for our subscribers Artificial intelligence: “France is far from being the only one to think that this is an issue of sovereignty”

In the early days of the Internet, to raise funds, many uninspiring start-ups, financed and inflated by myriads of venture capital firms, went public, and everyone speculated. After having pocketed the capital gains, these companies let them fall according to the cynical model of “pump and dump” [« pomper puis vidanger »]hence the collapse and bursting of the bubble.

In 2024, there are few IPOs for AI: eight, instead of the twenty-one expected in the first quarter, according to PitchBook. There was an average of 300 per year between 1980 and 2000. However, Astera Labs’ valuation of $11 billion marks a sign of recovery. This is the second largest valuation in AI after OpenAI. This company indeed provides the infrastructure that everyone needs in AI.

You have 39.15% of this article left to read. The rest is reserved for subscribers.

source site-30