Sam Altman's 2024 warning that people are "begging to be disappointed" by artificial intelligence may be the most important four words investors should remember as they evaluate AI-heavy portfolios. The OpenAI CEO made the remark while discussing the development of ChatGPT's next version, but it carries profound implications for how the market has priced in AI's future potential.

Why Are AI Stock Valuations So Stretched Right Now?

The past few years have been dominated by AI enthusiasm in financial markets. Companies like Nvidia, SpaceX, and countless others have seen their stock prices soar on the promise of artificial intelligence transforming entire industries. Many of these companies now trade at valuations that assume extraordinary growth over the next five years.

The problem is straightforward: when expectations become this elevated, even impressive real-world results can feel like a letdown. Nvidia, for example, trades at a price-to-earnings-growth ratio of just 0.63, which might look cheap on paper. But that valuation assumes the company will deliver specific levels of growth for the next five years. If actual performance falls short of those assumptions, investors could face significant losses.

What Did Altman Actually Say About AI Disappointment?

In 2024, as OpenAI worked on advancing ChatGPT, Altman cautioned that expectations for what the technology could accomplish had grown unrealistic. He observed that people were "begging to be disappointed" because their hopes for the chatbot's capabilities had become inflated. While Altman remains optimistic about AI's long-term potential for humanity, his warning suggests he recognizes a gap between what the technology can do today and what investors believe it will do tomorrow.

Altman

This distinction matters enormously. ChatGPT and similar AI tools have genuinely made many tasks easier. They can generate images, draft professional emails, and through agentic AI, handle multi-step processes that once required hours of human work. Yet these real accomplishments may not match the transformative breakthroughs that stock valuations are pricing in.

How to Evaluate AI Stock Risk in Your Portfolio

• Examine Valuation Assumptions: Before buying any AI-exposed stock, understand what growth rate is baked into its current price. If a company needs to grow 30% annually for five years to justify its valuation, ask yourself how likely that is.
• Diversify Beyond AI Hype: Consider balancing AI stocks with more traditional, value-oriented companies that have predictable business models and lower expectations already priced in.
• Watch for Reality Gaps: Pay attention to quarterly earnings reports and product announcements. When companies miss expectations, even slightly, AI stocks have shown they can fall sharply because the margin for disappointment is so thin.
• Remember Technology Moves Fast: The tech industry has a history of rapid shifts. What seems like a sure bet today can become obsolete quickly, so avoid concentrating too heavily in any single AI narrative.

The danger Altman is highlighting is not that AI won't be important. Rather, it's that the market may have gotten ahead of itself in pricing in AI's benefits. When reality doesn't match expectations, the correction can be painful.

Meanwhile, Altman himself has been busy with other ventures. On June 16, 2026, he was scheduled to attend a major announcement in Tokyo where SoftBank launched a cybersecurity service powered by OpenAI technology. However, he appeared only in a short video message, explaining that his baby daughter was born earlier than expected. Mark Chen, OpenAI's chief research officer, attended in his place.

The SoftBank partnership highlights how OpenAI is expanding beyond consumer chatbots into enterprise security applications. SoftBank and OpenAI established a 50-50 joint venture called SB OAI Japan last year to develop and exclusively market AI services for the Japanese market. The new patching service targets Japan's top 3,000 companies that operate critical infrastructure, including airports, power systems, and transportation networks.

SoftBank CEO Masayoshi Son described Japan's vulnerability to cyberattacks as "a crisis," comparing modern threats to "machine gun" assaults rather than the "rifle shots" of the past. The service works by first diagnosing security weaknesses and then analyzing what patches are needed to close vulnerabilities. Everyone who attended the Tokyo presentation was offered a free diagnostic.

This real-world application of AI technology underscores an important reality: AI is delivering genuine value in specific domains like cybersecurity and infrastructure protection. The question for investors remains whether these tangible benefits justify the astronomical valuations many AI companies now command. Altman's warning suggests they should think carefully before assuming they won't be disappointed.