AI Bubble Debate: Why Investors Pour Billions Into AI Amidst Warnings

Reddit users question if AI capital flows stem from FOMO or genuine value, with discussions including:

• Comparisons to dot-com/railway manias (survivors rebound like Amazon)
• NVDA’s P/E (28.25 vs Costco’s 45.66) as evidence against bubble claims
• Views of riding the bubble for profits, gold rush analogy (picking winners), and multi-year bubble expectation due to infrastructure demand

• 2025 AI VC funding: $89.4B (34% of total VC, 18% of deals) with concentrated capital [5]
• Polarized valuations: foundation models (50-100x revenue), enterprise AI (20-40x) [5]
• Burry’s $1B short bet on NVDA ($187.6M) and PLTR ($912M) [1]
• US private AI infrastructure investment: $1.35T [2]
• AI capex projected to rise from $88B (2025) to $143B (2026) [3]
• 54% of fund managers see AI as a bubble (up from 41% in Sept 2025) [1]

• Split between bubble warnings (Burry, 54% fund managers) and industrial transformation (capex growth, strong EPS for leaders) [1,3]
• Reddit’s P/E comparison aligns with research on polarized valuations [5]
• Both sides agree on sustained capital flows but differ on long-term viability

• Risks: Bubble burst catalyst (unknown), overvaluation of foundation models, concentration risk
• Opportunities: AI infrastructure (NVDA), enterprise AI (lower valuations), post-bubble survivors

[1] Yahoo Finance: ‘The Big Short investor bets $1 billion against AI bubble’
[2] Economic Times: ‘Palantir CEO sends AI bubble warning but US private investments hit $1.35T’
[3] Seeking Alpha: ‘The AI Bubble Everyone Fears Doesn’t Exist’
[4] Investing.com: ‘Breakout quarter expected for Nvidia’
[5] AINvest: ‘Thinking Machines’ $50B Valuation Ambition’