With the Middle East dominating headlines, the AI boom has taken a back seat. However, it remains a fundamental pillar supporting the S&P 500 and Nasdaq indices. Once geopolitical tensions subside, questions about whether the huge spending on AI is really paying off or whether progress is starting to slow down could resurface, and for now, the picture remains somewhat contradictory.
Starting with adoption, a Harvard Business Review study found that AI use is growing, especially in the technology and financial sectors. The problem is that, although employees continue experimenting with AI tools and deep integration into workflows, the real economic payoff remains limited.
Then there’s the issue of trust due to a phenomenon known as “model collapse.” In simple terms, AI systems risk being trained on content generated by other AI systems rather than on human-created material. Early generations of neural networks learned from vast amounts of human writing, which gave their outputs depth and nuance. But as AI-generated content floods the internet, models are increasingly feeding on their own output, creating a feedback loop that could gradually degrade quality.
And it would be one thing if the mistakes were limited to text.
Amazon recently experienced a series of outages affecting its website and apps, some of which were linked to the use of AI-assisted coding tools. One outage lasted six hours, leaving customers unable to make purchases or even check product information. In response, according to the Financial Times, Amazon gathered engineers for a deep dive into the causes and plans to tighten oversight of AI-generated code changes, including mandatory reviews by more experienced engineers.
By the way, a 2025 Google DORA report found that although about 90% of developers already use AI in their work, only 24% say they trust AI “a great deal” or “a lot,” while 30% say they trust it “a little” or “not at all.”
Now, let’s assume AI doesn’t “get dumber,” and its usefulness continues to grow — even if primarily as a supportive tool that boosts productivity rather than replacing human judgment. What other risks remain?
Competition.
For example, OpenAI still leads the chatbot space, but competitors are catching up fast. By the time of an OpenAI IPO, the market could already be much more crowded, making the stock less appealing. Strong competition could also put pressure on companies that have invested heavily in the technology.