Wall Street Breakfast Podcast: Oracle's AI Warning Shot

Oracle - Oracle reported adjusted earnings of $2.26 per share, exceeding the consensus estimate of $1.64, while revenue increased by 14% to $16.06 billion, slightly below expectations [3] - Cloud revenue, which includes infrastructure and applications, rose by 36% to $8 billion, narrowly missing estimates; infrastructure revenue surged by 66% to $4.1 billion, while application revenue grew by 11% to $3.9 billion [3] - A key metric, remaining performance obligations (RPO), surged by 438% to $523 billion, driven by new commitments from major clients like Meta and NVIDIA [4] - Oracle recorded a $2.7 billion pre-tax gain from selling its stake in Ampere, which contributed positively to both GAAP and non-GAAP EPS [5] - The company is shifting towards "chip neutrality," indicating a strategy to work with various CPU and GPU suppliers while continuing to purchase Nvidia GPUs [5] - Analysts noted that Oracle's revenue miss raises concerns about margin compression as revenue growth may slow down [6] SpaceX - Elon Musk confirmed that SpaceX is preparing for a potential IPO, which could occur as early as mid-2026, targeting a valuation between $1 trillion and $1.5 trillion [6][7] - The IPO is expected to support over $30 billion in capital needs for AI-driven projects, including space-based data centers and lunar satellite factories [7] - Musk emphasized that SpaceX has been cash-flow positive for years and primarily uses buybacks for liquidity, with valuation tied to progress on Starlink and Starship [8] Meta - Internal conflicts are emerging within Meta Platforms between the new superintelligence group led by Alexandr Wang and long-standing executives, including product chief Chris Cox and CTO Andrew Bosworth [9][10] - Wang advocates for prioritizing advancements in AI to catch up with competitors like OpenAI and Google, while other executives focus on integrating AI into existing products [11] - Disputes have arisen over resource allocation, with competing teams vying for computing resources to either enhance social media ranking or train AI models [12] - Reports indicate a request for a $2 billion budget cut from Reality Labs to support Wang's team, although Meta's spokesperson disputed this characterization [13]