Image via Tom Tunguz
Friday, February 20, 2026
Microsoft's $50B bet to fix AI's inequality problem
Microsoft's doubling down on closing the global AI divide with a massive $50 billion pledge through 2030 (bold move), while Sam Altman's simultaneously demanding urgent international AI regulation via an IAEA-style body because apparently the stakes are now *that* high. Meanwhile, the rivalry between OpenAI and Anthropic just escalated from tense Super Bowl ad wars to CEOs literally refusing to hold hands at a summit (yikes), and if that wasn't enough, a Microsoft Copilot bug just exposed confidential emails despite security controls, which is a pretty wild reminder that we're shipping this stuff while the safety questions are still very much unsettled. Oh, and The Zvi's reporting that AI productivity gains are already visible while some folks are now talking 1-3 year AGI timelines instead of decades. So real talk: if we're 1-3 years from AGI, should we actually be worried about which CEO shakes whose hand?
Image via Tom Tunguz
Top Stories
Microsoft pledges $50 billion by 2030 to bridge the AI adoption gap in the Global South through infrastructure, skills training, multilingual capabilities, and locally-relevant solutions, recognizing that unequal AI access threatens to perpetuate global economic inequality.
Sam Altman urges the world to establish urgent AI regulation through an international coordination mechanism, warning that unmanaged risks from powerful open-source models and concentrated control of AI technology pose existential threats requiring society-wide defense strategies.
Fortune
Former colleagues Sam Altman and Dario Amodei are locked in an increasingly public rivalry, with recent Super Bowl ads and on-stage tension highlighting the bitter competition between OpenAI and the rapidly ascendant Anthropic.
Microsoft's Copilot Chat exposed confidential labeled emails due to a bug, prompting industry experts to warn that rapid AI deployment without adequate governance makes data leakage incidents inevitable.
The Zvi
Frontier AI capabilities are advancing faster than expected with productivity gains already materializing, while safety concerns mount over open models, military applications, and the five-year window to solve alignment before potential AGI deployment. Industry insiders increasingly expect transformative AI within years rather than decades.
Keep Reading
Enjoyed this issue?
Get daily AI intel delivered to your inbox. No fluff, just the stories that matter.