Some people today worry about an AI bubble and wonder if it might repeat the dot-com crash, but they’re unaware of the following facts: 1. Leading AI companies currently have P/E ratios of only 20 to 30 times—modest, even low. During the bubble, Cisco, the market leader, had a P/E of 150 times—and it wasn’t selling scarce computing power, but routers and network cables. 2. Leading AI companies are continuously setting new revenue records, and human demand for AI will only grow. We’re still at the very beginning—adoption rates in transforming industries and boosting productivity remain very low. Can you imagine going back to a world without AI? You can’t. 3. AI has only been around for a few years; hardware deployment isn’t even complete yet. Software breakthroughs, application explosions, and productivity surges are still ahead. The future holds robotics, space exploration, pharmaceuticals, and quantum computing—all deeply tied to AI’s advancement. The hallmark problem of the 2000 bubble was that countless companies had no profitability, and telecom infrastructure was massively overbuilt—fiber optics, switches, and network capacity far outpaced real demand. Today, the situation is exactly the opposite: computing power, storage, bandwidth, inference costs, model capabilities, and application scenarios are all expanding rapidly. Many demands don’t lack potential—they simply lack infrastructure that is cheap enough, abundant enough, and good enough. Mindlessly comparing today to a bubble—or bearishness based solely on price increases—is the thinking of uninformed observers who’ve missed the opportunity, born from toxic speculation.