Nvidia's Lukewarm Forecast Sparks AI Slowdown Fears | Bloomberg Talks

Nvidia's Q2 Performance and Q3 Outlook

• 📊 Nvidia's second quarter revenue beat estimates at $46.7 billion, with gaming revenue also exceeding expectations.
• ⚠️ However, data center revenue for Q2 was $41.1 billion, slightly below the estimated $41.29 billion.
• 📉 The company provided a tepid revenue forecast for the current period, projecting around $54 billion, which has fueled concerns about a potential AI spending slowdown.

Analyst Reactions and Market Concerns

• 📉 Analysts expressed that while the results might be considered good for a normal company, they fall short of the heightened expectations set by Nvidia's previous performance.
• 💰 A $60 billion share buyback was announced, which is seen as a sign of the company's strong free cash flow but also raises questions about capital deployment if growth is decelerating.
• ⚠️ Concerns are rising that the pace of AI investment might be unsustainable, leading to a potential compression of Nvidia's valuation multiples.

Challenges in the China Market

• 🚫 Nvidia reported no H20 sales to China-based customers in the second quarter, impacting its revenue.
• ❓ There is significant uncertainty regarding future sales to China due to US export restrictions and Beijing's policies, with analysts having a wide range of forecasts.
• 💡 The company did mention an $180 million release of previously reserved H20 inventory, sold to a customer outside of China, though this is a small amount relative to overall revenue.

Future Growth and AI Infrastructure

• 🚀 The Blackwell architecture saw revenue grow 17% sequentially, and gross margins are trending positively, with expectations to exit the year in the mid-70% range.
• ⚡ Despite the current concerns, the long-term AI investment cycle is expected to be substantial, with data center spend potentially reaching a trillion dollars.
• 🔌 Emerging constraints on data center growth include electricity availability and the reliability of first-generation systems for new architectures like Blackwell.

Broader Tech and AI Trends

• ☁️ Snowflake's strong earnings suggest that enterprise data is becoming crucial for specialized AI training, benefiting companies like Snowflake and MongoDB.
• 🏠 While AI boom concerns are present, significant investment in data centers by major tech companies like Meta continues, indicating ongoing commitment to AI infrastructure.
• 💡 The narrative is shifting towards companies training AI models on their own enterprise data, rather than solely relying on open internet data.