Core Viewpoint - The article discusses the potential bubble in the capital market driven by AI, questioning whether the current valuations are justified and how close the market is to a potential collapse [1][19]. Group 1: Market Sentiment - Optimists believe that today's investments will yield high returns in the long term, making larger valuations reasonable [1]. - Pessimists acknowledge the utility of AI but argue that current valuations are excessively high [1]. Group 2: Investment Positions - Michael Burry's hedge fund, Scion Asset, updated its holdings for Q3 2025, with short positions in Palantir and Nvidia making up 80% of its portfolio, exceeding $1 billion [3]. Group 3: Capital Expenditure Trends - The capital expenditure (Capex) in the tech sector has reached an 18% compound annual growth rate, nearing levels seen during the 1999-2000 internet bubble, indicating overheating in infrastructure investments [10]. - AI-related Capex includes investments in GPUs, AI chips, data centers, and energy, suggesting that computational power is becoming as critical as oil [10]. Group 4: Business Dynamics - A cyclical investment relationship exists among major players like Nvidia and cloud service providers, where revenue growth appears self-reinforcing despite no real change in value [13]. - The article references a report from Marathon Asset Management, drawing parallels between the current AI investment climate and the telecom bubble of the early 2000s, highlighting the risks of oversupply [17]. Group 5: Profitability Concerns - While some AI-native companies are generating profits, many large model companies are operating at significant losses, with reports indicating that OpenAI earns $13 billion but incurs losses of $20 billion [19]. - The article suggests that the high costs of talent and capital expenditures in AI may hinder profitability, especially in the domestic market where many companies struggle to generate revenue [19]. Group 6: Bubble Analysis - A UBS report indicates that the market may still be in the early stages of a bubble, as tech stock valuations are close to normal levels, and earnings growth remains strong [21]. - The potential for profit margins to decline as capital intensity increases and competition rises is highlighted as a risk factor [21].