/
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

Source: Microsoft plans to release an AI security tool this month that uses models from Anthropic, OpenAI, and itself, as a cost-effective alternative to Mythos

Microsoft is preparing to release a new AI security product, internally codenamed Project Perception, to capture a piece of companies' rising cyber defense spending.

The Information Aaron Holmes

Context & Ripple Effects

Related coverage shows Microsoft simultaneously deepening its use of Anthropic models and working to replace some third-party model usage with its own MAI models to lower costs. Anthropic’s Mythos has also been tested by the NSA for finding software vulnerabilities, giving the new product a clear comparison point in a security-sensitive use case.

Project Perception therefore extends Microsoft’s model-sourcing strategy into cyber defense: it can package external and internal models around an enterprise security workflow while competing on the cost of the resulting task.

First-order effects

  • Microsoft gains a planned security offering aimed at corporate cyber-defense budgets and positioned directly against Mythos on cost.
  • Anthropic and OpenAI become both model suppliers to Microsoft’s product and, in Anthropic’s case, adjacent competitors where the product overlaps with Mythos.

Second-order effects

  • Microsoft can use a mix of Anthropic, OpenAI, and MAI models to manage performance and inference costs, reducing the need for any one supplier to power the entire workflow.
  • Mythos and other AI-security offerings face pressure to demonstrate differentiated vulnerability-finding performance or pricing rather than relying on frontier-model access alone.

Third-order effects

  • If enterprises adopt multi-model security products, value may concentrate in the vendor that owns the security workflow, customer relationship, and cost controls—not necessarily the lab supplying the underlying model.
  • The move points toward AI-security procurement being evaluated on cost per useful security outcome; model providers may increasingly be interchangeable inputs unless they retain distinctive capabilities.

The trend: AI products are shifting from single-model showcases toward multi-model enterprise systems optimized for the cost and reliability of a specific business task.

Discussion

  • Aaron Holmes Aaron Holmes on linkedin
    SCOOP: Microsoft is overhauling its cybersecurity business to cash in on corporate anxieties about AI-powered hacks. …
  • @wildebees Wessel van Rensburg on bluesky
    Microsoft's answer to Mythos's cost problem: don't commit to one model.  A router switches between Anthropic, OpenAI, and Microsoft depending on task type.  Keeps costs down.  Also means Microsoft captures the security workflow regardless of which model wins the capability race. …