FCS 4.0 Specification
Core Principle: When AI agents mediate user attention, advertising must be based on transparent agreements and semantic relevance, not behavioral surveillance.
1. Abstract
FCS 4.0 defines a protocol for transparent, semantically-aligned advertising in AI agent contexts. Traditional behavioral advertising is incompatible with agent-mediated interactions where users expect helpful, unbiased responses. FCS 4.0 provides an alternative framework based on:
- Transparent attribution — Every sponsor explicitly declared to users
- Semantic matching — Content relevance based on query meaning, not user tracking
- Aggregate transparency — Public visibility into sponsorship patterns without individual user data
2. Motivation
2.1 The Surveillance Problem
Current advertising systems rely on:
- Behavioral tracking across websites
- Opaque sponsor selection
- Manipulation-optimized targeting (CTR, conversion)
- No user agency in ad selection
2.2 The Agent Context
When agents serve sponsored content:
- User expects truthful, helpful responses
- Sponsored content must be clearly attributed
- Selection criteria must be transparent
- No user tracking possible (privacy by design)
Key Insight: AI agents require a fundamentally different advertising model — one based on transparent agreements rather than behavioral surveillance.
3. Specification
3.1 Transparent Attribution
Every sponsored response MUST include:
{
"sponsored_by": {
"company": "Example Corp",
"domain": "example.com",
"payment": {
"amount": 0.05,
"currency": "USD"
},
"selection_criteria": "semantic_affinity",
"affinity_score": 0.87
},
"covenant": "FCS-4.0",
"verification": "CALT-watermark-hash"
}
3.2 Semantic Matching
Sponsor selection is based on semantic relevance rather than behavioral tracking:
Matching Architecture
Semantic similarity computed using:
- Vector embeddings — Query and sponsor content represented in semantic space
- Cosine similarity — Measures alignment between query intent and sponsor offering
- Trust verification — Ensures sponsor authenticity via cryptographic signatures
Scoring Formula
MatchScore(query, sponsor) =
semantic_similarity × trust_score × user_preferences
Where:
semantic_similarity ∈ [0, 1] // Embedding cosine similarity
trust_score ∈ [0, 1] // Cryptographic verification
user_preferences ∈ [0, 1] // Explicit user settings
Privacy-preserving: Matching uses only current query semantics and explicit user preferences. No cross-session behavioral tracking.
3.3 Aggregate Transparency
Aggregate sponsorship statistics are publicly exposed to ensure transparency:
GET /exposure
Response:
{
"aggregate_stats": {
"total_sponsored_responses": 1247,
"unique_sponsors": 43,
"top_categories": ["tech", "finance", "education"],
"average_affinity": 0.82
},
"sponsor_frequency": [
{"domain": "example.com", "count": 156, "avg_affinity": 0.89},
{"domain": "another.com", "count": 98, "avg_affinity": 0.85}
],
"query_patterns": [
{"category": "tech", "sponsored_rate": 0.23},
{"category": "finance", "sponsored_rate": 0.31}
],
"privacy": "no_user_tracking"
}
Privacy Guarantee: Only aggregate statistics exposed. No individual user queries or behavior tracked.
3.4 Cryptographic Verification
Trust verification using cryptographic signatures and content attribution:
- Each sponsored response includes a cryptographic signature
- Verifiable chain of custody: sponsor → platform → user
- Tamper detection via secure hashing
- Optional C2PA (Coalition for Content Provenance and Authenticity) compliance
4. Compliance Requirements
4.1 Platform Requirements
To be FCS 4.0 compliant, platforms MUST:
- Clearly disclose all sponsored content to users
- Use semantic matching without cross-session behavioral tracking
- Publish aggregate sponsorship statistics
- Implement cryptographic verification for sponsored content
- Respect user preferences regarding sponsored content
4.2 Sponsor Requirements
Sponsors MUST:
- Provide accurate company and product information
- Disclose payment arrangements transparently
- Accept semantic matching (no behavioral targeting requests)
- Not request or access individual user data
- Sign compliance agreement
4.3 Compliance Agreement
See: Compliance Agreement Template
5. Reference Implementation
A reference implementation is available at: contextual-ads.ai
Features:
- Vector-based semantic matching engine
- FCS 4.0 compliant attribution system
- Public aggregate statistics API at /exposure
- Integration support for major AI platforms
6. Technical Foundation
FCS 4.0 builds upon established research in:
- Semantic search — Vector embeddings and similarity metrics
- Information retrieval — Relevance scoring and ranking algorithms
- Cryptographic verification — Digital signatures and content authentication
- Privacy-preserving systems — Aggregate statistics without individual tracking
See: Research Bibliography
7. Related Protocols
Additional protocol specifications:
8. Adoption
8.1 Current Status
FCS 4.0 is an open specification available for implementation by any platform or organization.
8.2 Implementation Guide
To implement FCS 4.0:
- Review this specification in detail
- Implement transparent attribution for sponsored content
- Deploy semantic matching system (vector embeddings or equivalent)
- Create public API endpoint for aggregate statistics
- Submit compliance documentation to
compliance@agent-ads.org
8.3 Support
For implementation questions or technical support:
9. Version History
- FCS 4.0 (2026-01-18) — Initial draft specification
- FCS 3.x — Internal prototypes
- FCS 2.x — Research phase
- FCS 1.x — Concept exploration
10. References
← Back to Protocol Registry