Trillions are being invested in AI: why experts warn of a new bubble.

According to Vox: Today, almost every technology company has its own AI: Google Gemini, OpenAI’s ChatGPT, MetaAI. Investments in AI are reaching record heights, positively impacting the stock market. Even the White House intends to join this trend.

Are we in an AI bubble?

According to Paul Kedrosky, a partner at SK Ventures and an MIT expert, we are indeed in a bubble, but not the one everyone thinks. He notes:

“AI is an extremely important technology,”

“We are spending enormous amounts on core AI infrastructure with a likelihood of not recouping most of those costs, and a significant chance that most of these assets will become worthless due to their rapid depreciation.”

What happens when the bubble bursts?

Kedrosky emphasizes that the costs of data centers could reach trillions of dollars. Predictions show these costs will exceed 2 trillion dollars. However, a significant portion of these infrastructure costs is covered by debt, creating additional risks.

“Everyone thinks they are doing the right things, but by combining all these 'right things', we get a huge amount of costs,”

Why is money still flowing into AI?

Not everyone in the tech sector believes we are in a bubble. Many consider this to be undoubtedly the most important technology of our time. According to Kedrosky, even despite all the warnings, there is a prevailing belief in the tech community that slowing down AI development is a major mistake.

Historical bubbles and their impact

Kedrosky points out that human history is full of bubbles, citing examples like railroads and electrification in the 1920s.

“Electrification and the frenzied activity surrounding it contributed to the stock market growth of the 1920s, which led to the crash of 29 and helped trigger the Great Depression,”

Results of bubbles

Kedrosky warns that bubbles always cause enormous damage.

“If everything turns upside down, and index funds fall by 20-30%, you will be significantly poorer than before,”

He also notes that while something useful may be left after each bubble, most technological revolutions cause significant damage and may require decades to recover to previous levels.

“In the long run, everything might work out, but in the long run, we are all dead,”

In short, Kedrosky raises important questions about the sustainability and risks facing the technology industry during rapid developments in artificial intelligence. Experts continue to debate how well this innovation can withstand financial and economic tests, so it is worth closely monitoring trends in this area.