The Global Seal of Trust™
Certifying the AI that powers the trust layer.
AI assets that participate in or interact with the trust layer of the internet deserve proof, not promises. The Global Seal of Trust is The AI Lab's independent, per-asset certification programme across four categories: AI Models, AI Agents, AI Infrastructure, and AI Datasets.
Born in the TIP Protocol ecosystem, the Seal certifies the AI that classifies, signs, hosts, and trains the protocol itself, including third-party Classifiers, Verification Providers, Node Operators, and the datasets and agents around them. The same Seal is open to any AI asset, anywhere, that affects public trust. Each application triggers an independent audit. Passing assets receive a publicly displayable, globally verifiable seal and a permanent public listing on the AI Trust Registry.
Open to any organisation worldwide. AI Trust Council membership is not required, though members receive priority review and reduced audit fees.
TIP Protocol is the standard. The Global Seal of Trust is the audit.
TIP Protocol (Trust Identity Protocol) is the open standard The AI Lab authored for verified human identity and content provenance across the internet. It is the layer that answers two questions the modern web cannot: who made this, and what is it actually made of.
The Global Seal of Trust audits AI assets against the signals TIP Protocol produces. Start with the protocol, then certify the asset.
Verified Human Identity
A portable, post-quantum signed identity for creators, publishers, and operators. Proves who is behind a piece of content or an AI asset without surrendering personal data to platforms.
Content Provenance
Every article, image, dataset, or model artifact carries a declared origin: Original Human, AI-Assisted, AI-Generated, or Mixed. Honest labelling is rewarded, deception is detectable.
Deterministic Trust Score
A four-component score computed by the protocol, not opinion. Feeds the AI Trust Registry™ and is the input The Global Seal of Trust™ audits against.
Ready to see what we certify? The Global Seal of Trust covers four asset categories.
4 CERTIFIABLE ASSET CATEGORIES
AI Models
CLASSIFIERS. FOUNDATION MODELS. ORIGIN-DETECTION ENGINES.
AI models that classify, verify, or assess content authenticity and human-vs-AI origin. Includes third-party models operating as TIP Classifiers, foundation models that handle TIP-signed content, and any trained model whose decisions affect public trust.
AI Agents
AUTONOMOUS SYSTEMS. SIGNING AGENTS. CONTENT WORKFLOWS.
AI agents that act on or generate content the public will see, including agents that sign and register content with TIP, agents that integrate with the AI Trust Registry, and autonomous systems where origin and trust signals shape the outcome.
AI Infrastructure
TIP NODES. VERIFICATION PROVIDERS. AI COMPUTE.
The compute, signing, and operational infrastructure powering AI trust, including TIP Nodes, Verification Provider pipelines, federated DAG hosting, and the AI compute stacks that train and serve the protocol's classifiers.
AI Datasets
TRAINING CORPORA. CLASSIFIER DATASETS. BENCHMARKS.
Datasets used to train, fine-tune, or evaluate AI systems that affect content trust, including TIP Classifier training data, deepfake-detection benchmarks, provenance-labeled corpora, and any dataset certified for provenance, representational fairness, lineage integrity, and absence of harmful content.
THE THREE CERTIFICATION SEALS
Sentinel
Foundational Governance Certification
WHAT THIS SEAL PROVES
Certifies that this AI asset meets foundational governance and risk management benchmarks, including verified audit logs, documented risk surfaces, and a governance framework that withstands scrutiny from investors, partners, and regulators. For assets operating within the TIP Protocol ecosystem, the audit additionally verifies clean integration with the protocol and the AI Trust Registry.
AUDIT REQUIREMENTS
- >Risk Surface Mapping documentation verified
- >Audit log framework implemented and active
- >Governance policy structure documented
- >Data lineage and provenance trail documented
Guardian
Human-Centric AI Certification
WHAT THIS SEAL PROVES
Certifies that this AI asset operates in a demonstrably fair, explainable, and human-centric manner, verified against bias benchmarks, explainability standards, and data provenance requirements across all applicable demographics. For TIP-ecosystem assets, the audit aligns the asset's human-AI handling with the protocol's Origin Code taxonomy and verified-human standards.
AUDIT REQUIREMENTS
- >Algorithmic fairness audit across all demographics
- >XAI trace documentation verified
- >Bias mitigation protocols implemented and validated
- >Dataset representational fairness certified
Sovereign
Critical Infrastructure Certification
WHAT THIS SEAL PROVES
Reserved for AI infrastructure and high-stakes autonomous systems requiring zero-knowledge architecture compliance, real-time adversarial containment, and kill-switch protocol readiness. Within the TIP Protocol ecosystem, Sovereign is the certification level for primary Verification Providers, foundational Node Operators, and high-volume protocol-integrity classifiers.
AUDIT REQUIREMENTS
- >Zero-knowledge architecture audit completed
- >Kill-switch protocol benchmarks independently tested
- >Real-time adversarial containment demonstrated
- >AI Trust Council Expert Panel review and ratification
THE 5-STEP CERTIFICATION PROCESS
Apply
Submit your asset type, target seal level, and documentation through the portal.
Scope
Audit team scopes the assessment requirements, timeline, and preparation needed.
Audit
Independent auditors perform a comprehensive assessment of both technical and governance aspects.
Review
Results reviewed by the AI Trust Council Expert Panel with remediation guidance where applicable.
Certified
Passing assets receive a verifiable seal, a public listing on the AI Trust Registry, and an annual renewal schedule.