Strengthening Credibility Scoring with Real-Time Intelligence

Background

Vericore (TAO Subnet 70) is an AI verification engine designed to restore trust in how information is judged for news and prediction markets. The subnet focuses on credibility scoring, verification workflows, and intelligence derived from real-world data.

By analyzing public signals, sentiment strength, and contextual evidence, SN70 aims to improve how truth, reliability, and confidence are measured across the Bittensor ecosystem.

At Desearch (SN22), our mission is simple: make real-time, high-quality information accessible across Bittensor so agents and subnets can think, react, and evolve with fresh data.

This partnership connects SN70's credibility framework with SN22's decentralized real-time search infrastructure.

Challenge

As SN70 expanded its credibility scoring system, several structural requirements became clear:

1. Real-Time Signal Dependency

Credibility evaluation cannot rely solely on static datasets. It requires access to live discussions, sentiment signals, and emerging narratives.

2. Contextual Depth Beyond Replication

Simply reproducing publicly available content does not create meaningful intelligence. Miners need tools to extract contextual awareness and signal strength.

3. Scalable, Decentralized Data Access

A centralized data source would introduce fragility and reduce interoperability across subnets.

4. Incentive Alignment for Quality Outputs

SN70 needed a mechanism to reward miners for deeper analysis rather than surface-level responses.

Without access to structured, real-time social and web signals, credibility scoring would remain limited in depth and adaptability.

Solution

The solution was optional integration of Desearch APIs into SN70's miner and scoring workflows.

Real-Time Social Signal Integration

Through Desearch, SN70 miners can enrich their outputs using:

• Real-time Reddit discussions
• X (Twitter) search data
• Public opinion and sentiment signals
• Web-based contextual results
• Structured JSON outputs for scoring workflows

This enables miners to incorporate live social discourse into credibility evaluations.

Instead of replicating static content, miners can now analyze:

• Sentiment polarity
• Discussion intensity
• Topic momentum
• Cross-source verification signals

This strengthens contextual intelligence while preserving decentralization.

Optional, Incentive-Driven Architecture

The integration remains optional.

Miners are not required to use Desearch, but those who enrich outputs with deeper signal-based intelligence can be rewarded with higher rankings and stronger scores.

This maintains subnet flexibility while aligning incentives toward higher-quality credibility assessments.

Console-Level Integration

The partnership also extends in the opposite direction.

SN70 services will become accessible through the Desearch console, allowing users to leverage Vericore capabilities directly within their workflows.

Through the console, users will gain access to:

• Advanced credibility checks
• News verification workflows
• Credibility scoring signals
• Intelligence-based ranking support

This creates a two-sided integration model:

SN70 gains distribution and real users.

Desearch users gain access to verification infrastructure.

Conclusion

The SN22 × SN70 partnership represents a practical integration between decentralized search and credibility scoring systems on Bittensor.

SN70 focuses on restoring trust in how information is judged.

SN22 provides the real-time social and web data layer that makes deeper evaluation possible.

Together, they are building infrastructure that:

• Encourages higher-quality miner outputs
• Introduces live social intelligence into credibility workflows
• Strengthens cross-subnet interoperability
• Expands real-world adoption of decentralized verification systems

As integration progresses, this collaboration lays the groundwork for more robust, signal-aware credibility systems across the Bittensor ecosystem.

Connect with Vericore

Website | vericore.ai
X | @Vericore_S70