In a recent discussion on Goldman Sachs' Exchanges podcast, Jim Covello, a prominent skeptic of artificial intelligence (AI) investments, expressed growing concerns about the profitability of AI companies. Covello has been debating the financial viability of AI for nearly four years, and he believes that the timeline for achieving profitability is extending rather than contracting. He emphasized that at some point, companies must start making money, and if the narrative remains that it is still 'early' in the adoption of AI two years from now, it could signal a significant challenge for the industry. Covello's remarks come as major AI firms like OpenAI and Anthropic are approaching initial public offerings (IPOs) with valuations nearing $1 trillion, yet neither is currently profitable. He pointed out that the rapid advancement of AI technology has not translated into financial success for these companies. In fact, he noted that many businesses are incurring greater losses while trying to implement AI solutions compared to two years ago. This trend raises questions about the sustainability of the current investment climate in AI. During the podcast, Covello discussed the unusual dynamics within the supply chain, where semiconductor companies are thriving at the expense of other sectors. He highlighted a shift in his investment strategy, now favoring hyperscaler stocks over traditional semiconductor stocks, as he believes that hyperscalers will benefit in most scenarios. Covello's analysis suggests that the economic benefits of AI are not reaching the workers as expected, indicating a disconnect between corporate expectations and the reality of AI implementation. Furthermore, Covello pointed to a significant data readiness problem that hampers the deployment of AI technologies. He argued that the bottleneck in enterprise AI is not the quality of AI models but rather the underlying data infrastructure, which is often outdated and not designed to support the demands of autonomous agents. This situation complicates the path to realizing the full potential of AI, as organizations struggle with siloed systems and inconsistent data governance. Covello concluded that while the eventual payoff of AI is still possible, the current trajectory raises serious concerns about the industry's future profitability and sustainability.