Who is building the agent operating system?
Twelve dimensions across seven providers. Re-graded as announcements land, with the market's reaction shown alongside.
Google · Agent Platform Potential
Future control-plane potential · updated 2026-06-01
Android + Assistant + Cloud + data is the most complete raw hand to build an agent OS from.
If agents become the primary way people get things done, how well-positioned is this provider to be the platform they run on? The distribution, the runtime, the permissions, the trust.
- AnalysisApple Intelligence could bring a better Gemini experience than Google doesAppleInsider · 2026-06-08
Ming-Chi Kuo's thesis that Apple's ecosystem and UI could beat Google's own Gemini delivery, but Gemini sets the ceiling.
- MarketApple WWDC 2026 Preview: Major Siri Upgrade. Has Apple's AI Payoff Arrived?TradingKey · 2026-06-06
Market preview citing Goldman ($340) and JPMorgan; argues Apple's edge is controlling AI distribution to 2B+ devices.
- AnalysisWith Gemini 3.5 Flash, Google bets its next AI wave on agents, not chatbotsTechCrunch · 2026-05-19
Google pivots from conversational chat toward autonomous, multi-hour agentic task execution.
- NewsGoogle flips Antigravity into an agentic dev suite9to5Google · 2026-05-19
Antigravity 2.0's agent-first architecture for orchestrating multiple coding agents asynchronously.
- AnalysisFrom 'catch up' to 'catch us': How Google quietly took the lead in enterprise AIVentureBeat · 2026-04-22
Google moved from catch-up to leading enterprise AI via Gemini Enterprise, Vertex, and a full-stack agent platform.
- NewsGoogle rebrands Vertex AI into the Gemini Enterprise Agent PlatformTechCrunch · 2026-04-22
Google's answer to AWS Bedrock AgentCore and Microsoft Foundry for building agents at scale.
- PrimaryOne Year of MCP: November 2025 Spec ReleaseModel Context Protocol · 2025-11-25
MCP's rise from open-source experiment to de facto standard adopted across OpenAI, Microsoft, GitHub, and Google.
- AnalysisChecking In on AI and the Big FiveStratechery · 2025-06-23
How Apple, Google, Meta, Microsoft, and Amazon are differently positioned across infrastructure, models, and distribution.
- AnalysisAI Engineering Goes MainstreamLatent Space · 2025-06-13
AI engineering matures into a vertical; agents, reasoning, and orchestration tooling are the practical frontier.
- AnalysisThe Agentic Era: Orchestration and Distribution, Not Model Power, Define AI MoatsDecoding Discontinuity · 2025-05-02
As frontier models converge, advantage shifts to orchestration quality and distribution reach over raw benchmarks.