/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

Anthropic's data shows software engineering accounts for ~50% of its AI agent tool calls; the remaining verticals are greenfields most founders are overlooking

Anthropic's new data shows software engineering dominates agentic AI.  For founders, that's not a warning.  It's a treasure map.

Garry's List Garry Tan

Context & Ripple Effects

Anthropic's usage data makes software engineering the clearest current proving ground for agentic workflows, while leaving a long tail of less-developed use cases. Later coverage similarly characterized coding agents as daily drivers for highly paid professionals, reinforcing why coding has become the reference market for agent products.

The concentration matters because it distinguishes demonstrated demand from opportunity spaces that remain less validated. It also arrives as Anthropic's position in initial AI-tool purchasing was reported to be strengthening in company spending on AI tools, giving its usage patterns added commercial relevance.

First-order effects

  • Founders and product teams get a concrete signal to treat software engineering as the most crowded agent category, while evaluating non-coding workflows as potential areas for differentiated discovery.
  • Anthropic gains evidence that its agent tooling is used heavily in a repeatable professional workflow, even though the data does not establish that the same mix applies across the broader AI market.

Second-order effects

  • Agent startups targeting non-engineering work will need to demonstrate workflow-specific value rather than rely on coding-agent benchmarks; buyers may compare them against the reliability and daily usage already associated with coding agents.
  • As coding becomes the anchor use case, vendors serving professional-services workflows face pressure to package agents around domain tasks and human review, particularly as AI reshapes entry-level work in those fields.

Third-order effects

  • If adoption broadens beyond engineering, agent competition may shift from general-purpose model capability toward embedded, domain-specific workflow products with measurable operational outcomes.
  • The pattern points to workplace-agent generalization, but the current evidence is usage from one provider rather than proof that every vertical will adopt agents at the same pace.

The trend: Agentic AI is moving from a coding-led beachhead toward contested, workflow-specific adoption across professional work.

Discussion

  • @handotdev Han Wang on x
    what I would be working on if I started another company today [image]
  • @garrytan Garry Tan on x
    Software engineering accounts for nearly 50% of all AI agent tool calls. Healthcare, legal, finance, and a dozen other verticals are barely touched, each under 5%. That's a hundred AI unicorns waiting to be built. https://garryslist.org/... [image]
  • @levie Aaron Levie on x
    This chart is a good reminder of how much opportunity there is in AI agents right now. There will be plenty of horizontal opportunities for agents, but equally many workflows that need deep domain expertise to actually make the user successful at automating the unique processes […
  • @anthropicai @anthropicai on x
    New Anthropic research: Measuring AI agent autonomy in practice. We analyzed millions of interactions across Claude Code and our API to understand how much autonomy people grant to agents, where they're deployed, and what risks they may pose. Read more: https://www.anthropic.com/…
  • @anthropicai @anthropicai on x
    Software engineering makes up ~50% of agentic tool calls on our API, but we see emerging use in other industries. As the frontier of risk and autonomy expands, post-deployment monitoring becomes essential. We encourage other model developers to extend this research. [image]