Format
Models
Sam Altman publicly ties OpenAI's exit path to safety readiness for the first time, deferring a $852B listing while Anthropic pushes toward mid-October. The compute landlord thesis continues regardless.
◆ Lena Park
Agents at Work
Three major rounds in three months — AIR, Glow, and Zenity — show investors betting that AI agents need their own security layer. The question is whether the guardrails can keep pace with the platforms they protect.
◆ Dana Ellison
Commerce
PYMNTS Intelligence data reveals AI has won the discovery war but lost the checkout battle, with marketplace gravity and merchant liability concerns keeping agent-led transactions at 3% of total volume.
◆ Tessa Vaughn
Infrastructure
Salesforce and Google Cloud are connecting their platforms through the Model Context Protocol at Dreamforce 2026, with Tableau enforcing governance at query time. The infrastructure layer is becoming a utility.
◆ Blair Hayes
Money & Markets
xAI is charging $30 a month for a service that costs at least $71 to host. The math is a bet on infrastructure arbitrage, aggressive subsidization, and the hope that token costs will fall fast enough to outrun the burn rate.
◆ Nolan Pratt
Policy
GrokBot, Meta's Muse, and Apple's Siri AI are shipping consumer AI agents under three different liability models — and there's no federal statute governing any of them. Connecticut enforcement starts in 16 days.
◆ Priya Nair
Agents at Work
The Cisco AI POD for Splunk brings agentic workloads into the enterprise data center — forcing CIOs to choose between the agility of managed cloud and the sovereignty of their own perimeter.
◆ Dana Ellison
Infrastructure
Four major vendors shipped standalone agent governance products within weeks. The pattern points to something bigger than feature announcements.
◆ Blair Hayes
Models
The $25.5B chipmaker raised $912M on the STAR Market, but its prospectus reveals a structural dependency that turns the sovereign AI narrative inside out.
◆ Lena Park
Agents at Work
The DoD is buying reasoning models. The enterprise is deploying autonomous agents. The safety standards for the leap between them don't exist yet.
◆ Dana Ellison
