CURRENT | Foundations

The Metaphor Problem

What "Platform Shift" Tells You to Build

By Nora Kaplan— March 4, 2026

Feature image for article: What "Platform Shift" Tells You to Build

At MWC this week, Qualcomm's CEO described an agent-centric future. Lenovo's EVP compressed the idea to three words: "apps to intent." Neither executive used the phrase "platform shift." Within hours, analyst notes had supplied it for them.

When an industry reaches for a metaphor nobody actually said, the metaphor is doing work. "Cloud" told companies to build for elasticity. "Mobile-first" told them to redesign for touch. Each encoded assumptions about where value would accrue, and organizations that followed early were rewarded. But those metaphors worked because the platforms held still. The agent transition may be the first one where that stops being true.

The Metaphor Problem

What "Platform Shift" Tells You to Build

At MWC this week, Qualcomm's CEO described an agent-centric future. Lenovo's EVP compressed the idea to three words: "apps to intent." Neither executive used the phrase "platform shift." Within hours, analyst notes had supplied it for them.

When an industry reaches for a metaphor nobody actually said, the metaphor is doing work. "Cloud" told companies to build for elasticity. "Mobile-first" told them to redesign for touch. Each encoded assumptions about where value would accrue, and organizations that followed early were rewarded. But those metaphors worked because the platforms held still. The agent transition may be the first one where that stops being true.

Nora Kaplan

Nora Kaplan, former collaboration platform product leader turned technology writer. Studied human-computer interaction and spent years designing tools for knowledge work. Now writes about AI agents, work transformation, and how enterprise software reshapes human capability at TinyFish.

What Ready Requires

The Three Infrastructure Problems Hiding Inside "Agent-Ready"

By Rina Takahashi— March 4, 2026

Feature image for article: The Three Infrastructure Problems Hiding Inside "Agent-Ready"

MCP has 97 million monthly SDK downloads. ChatGPT, Gemini, and Copilot have all adopted it. The connective plumbing for agents is getting standardized at a pace that suggests the hard part is over. Meanwhile, nearly half of organizations still can't make their own data searchable by the systems that plumbing is supposed to connect.

That gap between adoption velocity and actual readiness is where agent strategies go quiet. "Agent-ready" turns out to be several different problems that depend on each other in ways most strategy conversations skip entirely. Which layer you're actually stuck on matters more than the strategy deck.

What Ready Requires

The Three Infrastructure Problems Hiding Inside "Agent-Ready"

By Rina Takahashi— March 4, 2026

MCP has 97 million monthly SDK downloads. ChatGPT, Gemini, and Copilot have all adopted it. The connective plumbing for agents is getting standardized at a pace that suggests the hard part is over. Meanwhile, nearly half of organizations still can't make their own data searchable by the systems that plumbing is supposed to connect.

That gap between adoption velocity and actual readiness is where agent strategies go quiet. "Agent-ready" turns out to be several different problems that depend on each other in ways most strategy conversations skip entirely. Which layer you're actually stuck on matters more than the strategy deck.

What We're Reading

Qualcomm's CEO Declares 2026 the Year of the Agent

Worth reading for the specific language about agents replacing apps. The framing tells you where the narrative is headed.

Deloitte on Why Layering Agents onto Old Workflows Breaks Faster

Only 11% of organizations have agents in production. Deloitte's argument for why: value requires redesigning operations, and most teams skipped that part.

How the Model Context Protocol Became Infrastructure

MCP Lands on the Executive Agenda, Security-First

Seven Ways Multi-Agent Systems Break in Production

What We're Reading

Qualcomm's CEO Declares 2026 the Year of the AgentWorth reading for the specific language about agents replacing apps. The framing tells you where the narrative is headed.

Deloitte on Why Layering Agents onto Old Workflows Breaks FasterOnly 11% of organizations have agents in production. Deloitte's argument for why: value requires redesigning operations, and most teams skipped that part.

Quick links

How the Model Context Protocol Became Infrastructure

MCP Lands on the Executive Agenda, Security-First

Seven Ways Multi-Agent Systems Break in Production

Declarations and Data

"Year of the Agent" landed on 2026 with the subtlety of a venture capital press release. Some of the numbers back it up. Market projections are steep. Adoption curves look real. More than half of surveyed organizations now have agents running in production.

The interesting story lives in the gap between the forecast decks and the ops dashboards. Data infrastructure wasn't built for autonomous systems that need context on demand. Governance is scrambling to catch up with agents already live. The optimism and the friction are both true, at the same time.

Six data points, laid side by side, tell that story clearly.

Declarations and Data

"Year of the Agent" landed on 2026 with the subtlety of a venture capital press release. Some of the numbers back it up. Market projections are steep. Adoption curves look real. More than half of surveyed organizations now have agents running in production.

The interesting story lives in the gap between the forecast decks and the ops dashboards. Data infrastructure wasn't built for autonomous systems that need context on demand. Governance is scrambling to catch up with agents already live. The optimism and the friction are both true, at the same time.

Six data points, laid side by side, tell that story clearly.

Market Forecast

The Capital Thesis: $52 Billion by 2030

MarketsandMarkets sizes the AI agents market at $7.84B in 2025, growing to $52.62B by 2030. That's a 46.3% CAGR, and it sits at the conservative end of analyst estimates. The money is betting loudly. Production teams are the ones who have to cash the check.

Adoption Forecast

Gartner Sees Eightfold Agent Jump by 2026

From under 5% in 2025 to 40% of enterprise apps featuring task-specific agents by end of 2026. An eightfold jump in twelve months. Worth noting: Gartner itself warns against "agentwashing," where basic assistants get relabeled as agents. The same firm issuing the forecast is hedging the definition.

Production Reality

Majority of Organizations Now Run Production Agents

LangChain's 2026 State of AI Agents report puts 57% of organizations running agents in production, up from 51% last year. Large enterprises lead at 67%. Quality remains the top deployment barrier, cited by 32% of respondents. The momentum is genuine. So is the struggle to keep things working well.

Cancellation Risk

Four in Ten Agent Projects Won't Survive

Gartner predicts over 40% of agentic AI projects will be canceled by end of 2027. The culprits: costs that blow past estimates, business value that stays fuzzy after deployment, and risk controls too thin to catch failures before they compound. Shipping is one thing. Sustaining is another.

Data Readiness

Half of Organizations Say Data Isn't Agent-Ready

Deloitte's 2025 survey found 48% of organizations cite data searchability and 47% cite reusability as blockers to AI automation. Enterprise data architectures were built around ETL pipelines and warehouses. Agents need business context to act, and most data stacks weren't designed to provide it on demand.

Governance Gap

One in Five Has Mature Agent Governance

Deloitte's 2026 State of AI in the Enterprise finds only 20% of companies have a mature governance model for autonomous agents. The other 80% are assembling oversight frameworks while their agents are already live. Governance is trailing deployment by a widening margin, and the agents aren't waiting around.

Infrastructure Signals

The agent ecosystem produces announcements faster than anyone can evaluate them. New benchmarks, protocol updates, framework launches. Most of it is noise for teams trying to make actual build-or-wait decisions.

So here are six signals worth tracking. Each one pairs a maturity indicator with a concrete number. Some look reassuringly solid. Others carry caveats that tell you as much as the headline figure does. Read together, they sketch where the stack stands in early March 2026, which turns out to be a more interesting place than the keynote slides suggest.

Infrastructure Signals

The agent ecosystem produces announcements faster than anyone can evaluate them. New benchmarks, protocol updates, framework launches. Most of it is noise for teams trying to make actual build-or-wait decisions.

So here are six signals worth tracking. Each one pairs a maturity indicator with a concrete number. Some look reassuringly solid. Others carry caveats that tell you as much as the headline figure does. Read together, they sketch where the stack stands in early March 2026, which turns out to be a more interesting place than the keynote slides suggest.

Protocol Scale

MCP Crosses 97 Million Monthly SDK Downloads

Over 10,000 active servers now speak MCP, with native client support in ChatGPT, Claude, Gemini, and Copilot. Anthropic, OpenAI, Google, and Microsoft all back the protocol through the Agentic AI Foundation. The connective tissue of the agent stack is no longer theoretical.

Framework Shift

Microsoft Folds AutoGen Into a Unified Framework

AutoGen is now in maintenance mode. Microsoft merged it with Semantic Kernel into Agent Framework, with the Release Candidate shipping February 2026 and GA targeted this quarter. Meanwhile, the broader landscape is sorting itself: LangGraph owns stateful orchestration, CrewAI targets fast multi-agent prototyping. Distinct lanes, finally.

Browser Agents

Browser Use Hits 89.1% on WebVoyager Tasks

On WebVoyager's filtered 586-task subset, Browser Use leads open-source browser agents at 89.1% success. Playwright holds 45.1% adoption among QA professionals. The interesting pattern: AI reasoning layers are landing on top of traditional automation, pushing selector maintenance from 15–25% of effort down below 5%.

Agent Registry

First Community Census Counts 104,504 Registered Agents

The Universal Agent Registry's inaugural monthly report indexes 104,504 agents across 15 registries. Fetch.ai's AgentVerse holds the largest share at 34.8%. Text generation still dominates with 61,649 entries, but 19 distinct capability categories suggest the ecosystem is growing past its chatbot origins.

Open Governance

MCP Moves to Linux Foundation's Agentic AI Foundation

Anthropic donated MCP to the Agentic AI Foundation, a Linux Foundation directed fund co-founded with Block and OpenAI. Google, Microsoft, AWS, Cloudflare, and Bloomberg signed on as backers. Salesforce and Microsoft Dynamics 365 have already added native MCP support. The governance question is now distributed across 50+ enterprise partners.

Observability Metrics

Four Core Metrics Emerge for Agent Observability

Practitioner consensus is coalescing around success rate per task, average steps to resolution, tool call accuracy, and cost-to-value ratio of reasoning loops. Stanford CFRM research suggests structured evaluation can reduce production failures by up to 60%. The hard part: benchmark scores and production reliability still tell very different stories.

Past Articles

The Workforce Nobody Manages

Three million AI agents operate inside corporations. Walmart employs 2.1 million people across twenty countries and can ...

What You're Measuring and How You're Measuring It

Your agent passes testing, fails in production. Another completes tasks while taking paths you can't see are fragil...

When Agents Can't Find What's Right in Front of Them

A web scraper pulls product prices from a retail site every morning. One day it returns empty results. The prices are st...

What Happens When Context Windows Run Out of Room

An agent conversation runs long enough and you hit the wall—the model can't hold everything anymore. Compression discard...

Past Articles

The Workforce Nobody Manages

Three million AI agents operate inside corporations. Walmart employs 2.1 million people across twenty countries and can ...

What You're Measuring and How You're Measuring It

Your agent passes testing, fails in production. Another completes tasks while taking paths you can't see are fragil...

When Agents Can't Find What's Right in Front of Them

A web scraper pulls product prices from a retail site every morning. One day it returns empty results. The prices are st...

What Happens When Context Windows Run Out of Room

An agent conversation runs long enough and you hit the wall—the model can't hold everything anymore. Compression discard...