All things AI w @altcap @sama & @satyanadella. A Halloween Special. 🎃🔥BG2 w/ Brad Gerstner
TL;DR
OpenAI and Microsoft executives detail their restructured partnership, revealing a $130 billion nonprofit controlling OpenAI's public benefit corporation, exclusive Azure hosting until 2030 or verified AGI, and over $1 trillion in compute commitments justified by steep revenue growth and insatiable demand for AI capabilities.
🏢 Partnership Structure & Economics 4 insights
Microsoft holds 27% stake after $13.4B investment
Microsoft's ownership diluted from roughly one-third to 27% on a fully diluted basis after investing approximately $13.4 billion since 2019.
$130B nonprofit created with focused mission
The restructuring capitalizes the OpenAI nonprofit with $130 billion in stock, directing the first $25 billion toward curing disease and AI resilience initiatives including cyber defense and safety research.
Exclusivity terms run through 2030 or AGI verification
Microsoft retains exclusive rights to host OpenAI's 'stateless APIs' until 2030, while the revenue-sharing agreement and exclusivity provisions terminate early if AGI is officially verified by an expert panel.
Open distribution for non-core products
While frontier models remain Azure-exclusive, OpenAI is free to distribute open-source models, Sora, agents, and consumer devices on competing cloud platforms immediately.
🧠 AGI Governance & Timeline 3 insights
Expert jury determines AGI achievement
Verification of AGI triggers contract changes through a jointly selected panel of experts who must make a relatively quick determination when artificial general intelligence capabilities are achieved.
Divergent views on AGI proximity
Nadella stated on earnings calls that nobody is currently close to AGI and does not expect it soon, while Altman maintains higher conviction in rapid capability advancement and steep improvement trajectories.
Partnership designed for AGI transition
Both leaders emphasized that even if superintelligence emerged tomorrow, Microsoft would remain the preferred distribution partner, with the structure built to handle 'surprising twists and turns' in capability development.
⚡ Compute Constraints & Investment Scale 4 insights
Massive compute commitments versus current revenue
OpenAI has committed over $1.4 trillion in infrastructure spend across the next four to five years with Nvidia, AMD, Oracle, and Azure, despite current revenues being in the tens of billions.
Revenue growth constrained by supply, not demand
Altman claims OpenAI is 'well more' than $13 billion in revenue and growing steeply, stating that if the company had 10x more compute capacity, revenue would scale nearly proportionally.
Intelligence demand follows price elasticity curves
Altman argues that demand for compute is undefined without price context, noting that if intelligence costs dropped 100x, usage would increase more than 100x as new economic use cases become viable.
Universal compute shortage across the ecosystem
Both OpenAI and Microsoft report being severely compute-constrained, with Nadella confirming that Azure's growth would be higher if additional infrastructure capacity were available.
🌍 Nonprofit Mission & Market Failures 3 insights
Addressing areas where capitalism falls short
The nonprofit structure targets domains like automated scientific discovery and disease cures where pure market forces may not optimally serve human interests, alongside funding economic transition studies.
Largest nonprofit creation in history
The OpenAI nonprofit is positioned to become one of the world's largest charitable organizations, designed to capture value growth while the PBC secures capital necessary for aggressive scaling.
Defending against political capture
Altman emphasized the importance of directing resources toward AI resilience and health without the political biases that often weight down traditional nonprofit organizations.
Bottom Line
OpenAI and Microsoft have constructed a unique economic structure where AGI verification triggers partnership renegotiations, while betting that unprecedented compute investments today will unlock exponential revenue growth and fund the world's largest nonprofit for global health and safety initiatives.
More from BG2Pod
View all
ChatGPT – The Super Assistant Era | BG2 Guest Interview
OpenAI's Nick Turley reveals ChatGPT evolved from a planned one-month demo to a 900-million-user product by prioritizing long-term retention over short-term revenue, with future growth hinging on transforming the AI from a passive chat tool into a proactive super assistant capable of autonomous action.
AI Enterprise - Databricks & Glean | BG2 Guest Interview
Databricks and Glean executives argue that while 95% of enterprise AI projects currently fail, this reflects necessary experimentation in a market where LLMs have become commodities and true competitive advantage comes from leveraging proprietary data through learning-based systems rather than brittle automation.
AI Bubble, Stablecoin Boom, and Runnin' Down a Dream | BG2 w/ Bill Gurley and Brad Gerstner
Bill Gurley announces his departure as BG2 co-host to focus on policy work and his new book, while he and Brad Gerstner debate whether AI infrastructure spending constitutes a bubble, analyzing circular revenue transactions, unprecedented Big Tech capex levels, and the competitive dynamics driving potential overbuilding.
NVIDIA: OpenAI, Future of Compute, and the American Dream | BG2 w/ Bill Gurley and Brad Gerstner
Jensen Huang argues that AI compute demand will explode due to three compounding scaling laws (pre-training, post-training, and reasoning), while dismissing Wall Street fears of a coming glut by framing the shift from general-purpose to accelerated computing as a multi-trillion dollar infrastructure transition that is still in its early innings.