/
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

Microsoft adds automatic AI model selection to Visual Studio Code; it primarily favors Claude Sonnet 4 over GPT-5 or GPT-5 mini for paid GitHub Copilot users

Visual Studio Code now has a new auto AI model selector that favors Claude 4 over GPT-5.

The Verge Tom Warren

Context & Ripple Effects

GitHub Copilot Chat began as a GPT-4-powered feature inside Microsoft’s IDEs, making the coding assistant a core part of the developer workspace rather than a separate tool. VS Code’s new selector shifts that experience from a default-model product toward runtime choice based on the task.

The preference for Claude Sonnet 4 also fits Microsoft’s later direct integration of Claude and Codex coding agents across GitHub and VS Code, extending Copilot beyond its earlier GPT-centric positioning.

First-order effects

  • Paid GitHub Copilot users in VS Code will have model choice automated, with Claude Sonnet 4 receiving the primary preference over GPT-5 and GPT-5 mini.
  • Anthropic gains privileged default placement in a major coding workflow, while OpenAI’s GPT models become alternatives within that workflow rather than the apparent first choice.

Second-order effects

  • Model vendors must compete not only for raw coding quality but for selection criteria inside Copilot; default routing can materially shape which models developers actually use.
  • Developers gain less need to manually compare models, but also become more dependent on Microsoft’s routing logic to balance response quality, speed, and availability.

Third-order effects

  • Coding assistants are moving from a single-model feature to a managed multi-model layer, where the IDE provider owns the user relationship and allocates demand among underlying model suppliers.
  • If this pattern persists, differentiation will increasingly depend on measurable performance in specific workflow tasks and on distribution access, not solely on a model vendor’s standalone chatbot brand.

The trend: AI coding platforms are becoming multi-model orchestration layers that automatically route developer tasks to the model the platform judges most suitable.

Discussion

  • @tomwarren.co.uk Tom Warren on bluesky
    Microsoft favors Anthropic over OpenAI for Visual Studio Code.  Microsoft's code editor now has a new auto AI model selector that favors Claude 4 over GPT-5, which is something I've heard Microsoft has been favoring in recent months.  Full details 👇 www.theverge.com/report/77864.…