A viral chart comparing the S&P 500’s current trajectory with that of the late 1990s has reignited debate over whether enthusiasm for artificial intelligence is inflating a classic market bubble.
The index keeps setting records in 2026, and analysts disagree sharply on what that pattern actually signals.
The Warning Signs That Alarm the Bears
The Shiller CAPE ratio measures share prices against inflation-adjusted earnings over a decade, offering a longer view than conventional metrics. It currently hovers near 40. That level carries historical weight. Similar readings appeared only at the absolute peak of the dot-com bubble.
Analyst Rekt Fencer triggered the discussion. He posted a chart overlay arguing that the structures look almost identical: a sharp correction, a robust recovery, then a renewed push toward new highs.
His framing was deliberately provocative. He listed 1999 as the dot-com bubble, 2007 as the housing bubble, then asked whether 2026 represents the AI bubble.
“…The S&P 500 is trading at valuations not seen in 100 years. Retail demand remains near record highs. We saw the same combo before: 1999: Dot Com Bubble 2007: Housing Bubble 2026: AI Bubble The scariest part? The chart is now mirroring the Dot Com bubble almost perfectly…,” Rekt Fencer said on X.
Follow us on X to get the latest news as it happens.
The warning centered on complacency. Everyone thinks this time is different, he wrote, adding that the assumption itself is what worries him most. Similar posts have circulated widely on X. Various chart overlays suggest the market is tracing paths seen before previous major peaks.
Concentration data reinforces those concerns. Technology now accounts for roughly one-third of the S&P 500, with the top 10 companies comprising nearly 40%. Those levels exceed previous extremes. Both late 1999 and the mid-2000s showed lower concentration than markets display today.
Bubble proponents cite additional signals. Circular financing arrangements among AI players, massive data-center capital expenditure, and retail enthusiasm all echo late-stage patterns.
Why Others See 1997 Instead of 1999
The counterargument rests on fundamentals. Unlike many dot-coms that burned cash with minimal revenue, today’s AI leaders generate substantial profits.
Funding sources differ meaningfully. NVIDIA, Microsoft, and the hyperscalers finance expansion largely through free cash flow rather than speculative debt or endless equity issuance.
Valuation multiples support that distinction. Forward price-to-earnings ratios for technology sit around 25x to 30x, well below the 50x to 58x peaks of the early 2000s.
Goldman Sachs frames the setup differently. The bank argues conditions resemble 1997 more than 1999, with investment rising but extreme imbalances not yet present.
Amundi research reached similar conclusions. Its analysis found the 2023 to 2025 rally lacks the explosive valuation dynamics typical of late-stage bubbles.
Skeptics also question the charts themselves. Overlays can be selective, and genuine enterprise demand distinguishes this cycle from pure speculation. History offers ambiguous guidance. Transformative technologies from railroads to the internet produced both lasting value and painful interim excesses.
The resolution depends on earnings quality. Records alone settle nothing, and the durability of AI-driven growth will determine which parallel proves accurate.
One observation applies regardless of outcome. The phrase this time is different has proven both true and expensive, sometimes simultaneously.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights.
Bitrue