After explosive growth and hitting huge valuations, investors are now questioning whether the AI boom is sustainable or on the edge of a bubble. Although the pace of technology evolution keeps intriguing the markets, geopolitical tension, with the trade wars, as well as ethical concerns and a potential overestimation of industry applications for these new tools are raising concerns.
The enthusiasm surrounding AI has led to massive investments, particularly in data centers and chip manufacturing over the past decade. Spending on AI data centers between 2024 and 2027 is projected to exceed $1.4 trillion, according to a recent report from Economist Impact. Now, it’s the time for those companies to prove real generation of value for stakeholders and investors.
Valuations also reached scary numbers. Chipmaker Nvidia saw its market value increase to over $3 trillion, entering the select hub of trillionaire companies among big techs like Meta, Amazon, Alphabet (Google) and Apple. Initially designed to be an NGO, OpenAI is currently valued at $300 billion, something unimaginable a couple years ago, before being invested by Microsoft, which now holds a relevant stake in the ChatGPT creator.
To keep an eye on
Concerns are mounting over the sustainability of current AI investments. Training large AI models requires vast amounts of energy and data, leading to escalating costs. For instance, training GPT4 consumed enough electricity to power five thousand American homes for a year. Specialists on the field also mention a shortage of high-quality training data available, as well as judicial disputes regarding privacy and copywriting.
Others caution against an exaggerated optimism. Alibaba Chairman Joe Tsai has expressed concerns about overinvestment in AI, particularly in the US, where companies may be pouring funds into AI ahead of actual demand. This sentiment is echoed by experts who warn that the influx of money into generative AI without clear returns on investment has inflated expectations to unsustainable levels.
The stock market has also reflected these uncertainties. AI-related stocks, including major players like Nvidia, have experienced significant volatility in the past weeks. Despite strong earnings, these companies have faced stock price declines due to factors like the trade war and speculation. Yet, the long-term outlook for AI remains promising: advancements in multimodal models and automated machine learning are expected to democratize AI and open up new possibilities.
Not your old DotCom
Despite these challenges, some experts argue that the AI sector is not in a bubble. The main argument is built around the idea that leading AI companies are already highly profitable and have lower valuations when proportionally compared to the late 1990s, suggesting a more stable and solid foundation for these new companies.
While the AI sector faces significant challenges and potential overvaluation, it also holds substantial promise. The coming years will be critical in determining whether AI can deliver on its lofty expectations or if the current enthusiasm will wane. Investors and companies alike must navigate this landscape with caution, balancing optimism with actual results.
