The Fight Over Open Source AI, Anthropic's $1.5B Payout, NYC Socialists: Evictions = Violence?

Summary

The hosts debate the rapid rise of Chinese open-source AI model Kimi K3, which matches US frontier labs but is cheaper, sparking calls to ban Chinese open-source. David Sacks argues against a ban, calling it a tragic mistake that would backfire, while Chamath Palihapitiya warns that AI foundational models are commoditizing fast. The discussion also covers Anthropic's $1.5B copyright settlement, with Sacks exposing hypocrisy in its IP theft claims. Later, Chamath and David Friedberg make a strong bullish case for Google as the best AI public market stock after its earnings, while Jason Calacanis argues training data owners like the New York Times and Reddit will be the real winners.

• Kimi K3 matches top US models, reigniting the debate over banning Chinese open-source AI; David Sacks says a ban would hurt US developers and help China.
• Chamath Palihapitiya contends AI models are commoditizing, pushing value to the application layer and infrastructure, putting closed labs at risk.
• Anthropic settled a book piracy lawsuit for $1.5B; Sacks highlights the hypocrisy in calling Chinese distillation 'IP theft' while claiming fair use to train on others' content.
• Jason Calacanis believes training data owners (NYT, Reddit) will be the biggest beneficiaries as AI models become commodities.
• Google's stock fell on surging AI capex, but Chamath and Friedberg argue the company's 32% ROIC, model-agnostic cloud, and low-cost infrastructure make it the top AI public market play.
• Tesla also reported a capex surge and negative free cash flow, with no strong investment call from the hosts.
• In 'Socialism Corner,' the hosts criticize NYC policies equating evictions with violence, warning that eroding private property rights leads to tyranny.