DeepSeek Becomes the Kill Line for LLMs
Will DeepSeek's rock-bottom model pricing eliminate OpenAI and Anthropic? The answer is more complex than that.
An investor’s view for founders: industry research on HK, US, and A-share markets, sector analysis of AI and semiconductors, quantitative and trading approaches, and asset allocation thinking on the path to financial independence (FIRE).
Will DeepSeek's rock-bottom model pricing eliminate OpenAI and Anthropic? The answer is more complex than that.
July 2026 saw AI trading turn risk-averse, sinking chips, memory, and software stocks as the Nasdaq 100 fell 9.5% from its high.
Situational Awareness Fund sells most of its public equity portfolio to Citadel, surrendering a successful AI trade amid market reversal.
Against the backdrop of U.S. stock AI giants reveling in valuations soaring by tens of multiples, Nvidia's Jensen Huang has surprisingly done something.
Over two years, AI infrastructure focus shifted from GPUs to power, then optical modules, liquid cooling, transformers, and data center REITs.
Over the past two years, market discussions on AI infrastructure have primarily focused on three key aspects: First, who has the GPUs.
Investor Serenity (@aleabitoreddit) goes viral on X with a 4502% YTD return and 500k followers gained in 4 months.
On May 22, 2026, eight Chinese agencies cracked down on illegal cross-border finance, issuing maximum penalties to brokers like Futu and Tiger.
A top AI investor, ex-OpenAI researcher Leopold Aschenbrenner, just updated his portfolio—and the data reveals a surprisingly aggressive strategy.
In June 2024, Leopold Aschenbrenner published "Racing to the Trillion-Dollar Cluster," predicting the AI landscape two years ahead while others debated GPT-4.
Last year, I heavily invested in Alibaba, betting on its full AI stack. Now, I'm deeply disappointed by its performance.