China Tech Rout Worsens on AI Spending

🔑 Enhanced Key Takeaways

• •Chinese AI companies face a divergence in investor sentiment: while 'pure-play' AI firms like MiniMax Group saw stock values more than double in February 2026 due to regulatory protection from foreign models, broader tech megacaps are experiencing losses as investors reassess whether massive AI capex spending will generate sufficient returns[1][2].
• •Goldman Sachs Research reports that AI hyperscaler capital expenditure estimates for 2026 have reached $527 billion (up from $465 billion), but investor correlation across large public AI companies has collapsed from 80% to just 20% since June 2025, indicating selective skepticism about which AI investments will be profitable[2].
• •Chinese AI models have demonstrated that expensive NVIDIA chips may not be necessary for frontier model training, challenging the hardware-centric investment thesis that has driven much of the 2025 AI capex boom and potentially undermining returns for infrastructure-heavy players[4].
• •The BlackRock Investment Institute projects $5-8 trillion in additional AI-related capex through 2030, but regulatory risks including data collection scrutiny, potential fines, and country-specific regulations could significantly impact AI company valuations and profitability[5].

🔮 Future ImplicationsAI analysis grounded in cited sources

⏳ Timeline