Claude Usage Limits Doubled Through March 27 — All Tiers, All Surfaces
Anthropic is running a two-week promotion that doubles Claude usage limits across Free, Pro, Max, and Team plans. The bonus applies outside weekday peak hours (8am–2pm ET) and covers every Claude surface: claude.ai, Cowork, Claude Code, Claude for Excel, and Claude for PowerPoint. No settings to toggle — it’s automatic.
The timing is smart. With the Pentagon battle dominating headlines and millions of new users flooding in over the past two weeks, this is Anthropic’s way of converting trial users into habitual ones. Double the limits means double the chance someone hits the moment where Claude becomes indispensable. The promotion runs through March 27.
Advanced Tool Use Ships: Tool Search, Programmatic Calling, and Tool Examples
Anthropic introduced three new beta features that fundamentally change how Claude discovers and executes tools. Tool Search lets Claude search across thousands of available tools without loading them all into context — preserving 191K tokens compared to the traditional approach, an 85% reduction. Programmatic Tool Calling lets Claude invoke tools via a code execution environment instead of structured JSON. Tool Use Examples provides a universal standard for demonstrating how to use any tool.
This matters for anyone building agentic systems. The token savings from Tool Search alone change the economics of tool-heavy agents. Instead of cramming hundreds of tool definitions into every request, Claude can discover the right tool on demand. Combined with programmatic calling, this makes it practical to give Claude access to massive tool libraries without context window bloat.
Code Execution Tool v2: Bash Replaces Python-Only Sandbox
Anthropic launched v2 of the Code Execution Tool in public beta, replacing the original Python-only sandbox with full Bash command execution and direct file manipulation. This is a significant upgrade for developers building agents that need to interact with file systems, run shell scripts, or chain multiple command-line tools together.
The original code execution tool was limited to Python scripts in an isolated environment. v2 opens the door to much richer agent behaviors — think git operations, data processing pipelines, or system administration tasks. It pairs naturally with the new programmatic tool calling feature, creating a powerful combo for agentic workflows.
Enterprise Analytics API and Usage & Cost API Go Live
Two new APIs for org admins. The Enterprise Analytics API gives programmatic access to usage and engagement data for Claude and Claude Code Remote, aggregated per org per day. The Usage & Cost API lets administrators monitor spending programmatically, with a new Admin API endpoint for org info retrieval.
These are table-stakes features for enterprise adoption. Companies running Claude at scale need visibility into who’s using what and how much it costs. The fact that these are separate APIs (rather than just dashboard views) means teams can pipe the data into their existing FinOps and observability stacks.
Microsoft Making Claude Sonnet Available in M365 Copilot
On March 9, Microsoft announced it will make Anthropic’s latest Claude Sonnet models available to M365 Copilot users. This is a notable expansion of Claude’s enterprise footprint — Microsoft is effectively acknowledging that its own models aren’t the best fit for every task and giving enterprise customers the option to route certain workloads through Claude instead.
For Anthropic, this is distribution at massive scale. M365 Copilot is embedded in the daily workflow of millions of enterprise knowledge workers. Even a small percentage routing tasks to Claude represents significant volume and revenue. The move also further cements the trend of enterprise AI platforms becoming model-agnostic.
Anthropic Study: AI Hiring Impact on Younger Workers Is Real
Anthropic’s latest research found that hiring has slowed for younger workers in certain occupations due to AI’s growing influence. The study, covered by HR Dive, suggests the labor market impact of AI tools like Claude is no longer theoretical — it’s showing up in hiring data, particularly for entry-level roles in occupations with high AI exposure.
This is consistent with Anthropic’s earlier labor market study (March 8) that found 75% of programming tasks are AI-exposed but actual displacement remains low. The nuance matters: AI isn’t eliminating jobs wholesale, but it is reshaping which jobs get created and who gets hired for them. Junior roles that involve routine tasks are the first to feel the squeeze.
Judge Fast-Tracks Pentagon Hearing to March 24 as Anthropic Warns of Billions at Stake
The Anthropic-Pentagon legal battle escalated significantly. Anthropic told the presiding judge that it could lose billions in revenue this year from the supply chain risk designation and urged quick action on its request for an injunction. The judge agreed, moving the hearing up to March 24 — just eight days from now. In a joint letter, dozens of AI scientists and researchers from OpenAI and Google expressed support for Anthropic’s position, arguing the designation threatens the broader AI research ecosystem.
The coalition of support from rival companies is remarkable. OpenAI and Google researchers publicly backing Anthropic signals that the industry views the Pentagon’s action as a threat to all AI companies, not just one. If the government can designate a leading AI lab as a supply chain risk for refusing to drop ethical guardrails, every AI company faces the same exposure. The March 24 hearing is now the most consequential date on the AI policy calendar.
$20B ARR and 40% Enterprise Share: The Numbers Tell the Story
Anthropic was on pace for $9B in annual revenue at the end of 2025. By early March, that figure had reportedly nearly doubled to $20B. The share of US companies paying for Claude hit 20% in January, up from roughly 4% a year ago. Meanwhile, Anthropic’s share of enterprise AI spending climbed to 40%, while OpenAI’s fell from 50% to 27% over the same period.
These numbers are staggering for a company that was widely considered an underdog just 18 months ago. The revenue acceleration is driven by three things working simultaneously: Claude Code’s explosive growth ($2.5B run-rate), the consumer surge from the Pentagon controversy (11M daily users), and enterprise adoption accelerating through the Partner Network. OpenAI is still larger overall, but the gap is closing fast and the trajectory lines are clear.