Tae Kim argues AI is underhyped; hosts debate dot-com bubble parallels and console war end
Oct 27, 2025
Key Points
- Tae Kim argues AI remains underhyped with reasonable valuations and low leverage, positioning the market in early innings of a computing shift comparable to 1994 rather than 1999.
- Hosts counter that early-stage bubbles like the dot-com era can generate real wealth for early winners before eventual correction, making timing difficult to pinpoint.
- The debate hinges on whether current AI product momentum and valuation discipline prove we're in a sustainable growth phase or an early bubble destined to correct.
Summary
Tae Kim argues AI is underhyped and not yet in a bubble, pointing to concrete metrics rather than sentiment. Big tech valuations are reasonable, leverage is low, and multiple AI super-product cycles are arriving in the year ahead. He frames this as the early innings of a computing shift to AI, the largest in decades, and compares the current moment to 1994 rather than 1999 in the dot-com era.
The hosts question the timeline. Even large bubbles start small. Amazon went public in May 1997, yet the dot-com crash did not arrive until 2000–2001. By that logic, an AI bubble could already be in its very early stages without detection. They distinguish between the late 1990s, when euphoria and valuations spiraled, and the earlier period when early winners still existed and could exit profitably. One cites a founder who sold a company in 1998 for roughly $100M in liquid cash and never saw the company again after 2002, a successful exit before the collapse.
Kim's argument rests on valuation discipline and product momentum today. The hosts grant that early-stage bubbles can generate real wealth for early winners, even if the broader cycle eventually corrects. Whether we're in 1994 or 1999 determines the stakes, but neither side fully resolves the point.