What Is Biteris?

Biteris is a cryptographic registry and deterministic verification engine. It answers one question with mathematical certainty: Has this specific piece of data been altered since it was locked?

Unlike ChatGPT or Gemini, Biteris doesn't guess, summarize, or generate opinions. It compares cryptographic hashes and returns either "Match" or "No Match." This means it cannot hallucinate—because hallucination requires interpretation, and Biteris does none.

The Human Verification Gate

Every single submission to the Biteris registry is verified by a human before it is cryptographically locked. This is the critical difference between Biteris and automated systems that accept whatever data is uploaded.

When a professional submits a credential, a trained verifier examines it against authoritative sources. Only after human confirmation does the system generate the cryptographic hash and store it on the distributed node network.

Consequence of malicious activity: Any user who attempts to submit fraudulent, forged, or otherwise malicious data is permanently banned for life from the Biteris ecosystem. No appeals. No second chances. This zero-tolerance policy creates a powerful deterrent against bad actors.

How It Works (In Plain English)

1. A professional submits their credential (e.g., "Dr. Smith, Medical License #TX123456")

2. A human verifier checks the submission against authoritative sources (state medical board, official database, etc.)

3. If verified, the system creates a unique digital fingerprint (hash) of that exact data

4. That fingerprint is stored across a distributed node network

5. When an employer queries the credential, Biteris generates a fresh fingerprint of the query

6. If the fingerprints match: Verified. If they differ by even one character: Not Verified.

No ambiguity. No "confidence scores." No plausible fictions. And critically: no automated acceptance of unverified data.

Current Scale (Realistic, Not Hype)

Metric	Current Status
Human verifiers	47 trained professionals across 12 time zones
Node distribution	122 nodes across 47 jurisdictions (enterprise/partner deployments)
Verified entities	~47,000 with active Global Trust Scores
Professional profiles	~8,000 software developers, ~3,500 healthcare professionals, ~2,200 attorneys
Peak capacity	1.4M verifications/second (load-tested, not sustained current traffic)
Active deployments	Pilot programs in 3 US school districts, 5 logistics firms, multiple Fortune 500 evaluations
Malicious submission rate	0.04% (and falling, as lifetime bans deter repeat offenders)
Lifetime bans issued	142 (as of June 2026)

What Biteris Does Not Do

• It does not automate verification. Every submission requires human review. This adds latency (typically 24-72 hours for initial verification) but ensures quality.

• It does not prevent human error in verification. Verifiers can make mistakes, though rigorous training and multi-reviewer sampling minimize this risk.

• It does not eliminate node collusion risk. In theory, a coalition of node operators could collude. (Mitigated through geographic distribution and audit trails.)

• It does not replace all trust with math. It's a verification utility with a human-in-the-loop, not a magic wand.

Who Is Using It (Early Adopters)

• Schools: Staff credential verification (pilot deployment in three US districts)

• Logistics: Cross-border escrow settlements (five active logistics firms)

• Professionals: Individual credential storage across 50 categories (biteris.net/top50)

Why This Matters

In an era of AI-generated hallucinations, deepfakes, and forged credentials, organizations need a way to verify claims without relying on the same probabilistic systems that create the noise. Biteris combines cryptographic certainty with human judgment: every locked credential has been seen, checked, and approved by a real person. And anyone who tries to cheat is banned for life.

Bottom Line: Biteris is a useful tool for a specific problem. It works today, has known limitations, and continues to grow through enterprise adoption—with human eyes on every single submission.

---

Full Technical Deep Dive (Whitepaper Length)

For readers who want the complete architecture, API specifications, and risk analysis.

Table of Contents

1. The Problem: Verification in an Age of Generative Noise

2. The Solution: Human-Verified, Cryptographically Locked Attestation

3. The Human-in-the-Loop: Verification Process and Lifetime Bans

4. Architecture Overview: Off-Grid, API-First, Node-Based

5. Cryptographic Primitives: Current and Roadmap

6. Real-World Use Cases with Current Status

7. The Top 50 Professions Framework

8. Known Limitations and Risk Factors (Transparency Section)

9. API Reference for Developers

10. Comparison with Alternative Approaches

11. Conclusion: A Tool, Not a Panacea

---

Part 1: The Problem – Verification in an Age of Generative Noise

The internet has a verification problem. Traditional search engines return results optimized for engagement, not accuracy. Large Language Models generate plausible-sounding text that is frequently incorrect—a phenomenon called hallucination. Credential fraud costs organizations billions annually. Automated systems, no matter how sophisticated, can be gamed.

These problems share a common root: the absence of a reliable, human-in-the-loop method to verify that a specific claim about a specific entity is both true at origin and unchanged since.

Biteris addresses this specific problem and nothing else—by putting a human at the front door.

---

Part 2: The Solution – Human-Verified, Cryptographically Locked Attestation

Core Principle

Automated systems accept whatever is uploaded. Biteris does not.

Every submission is verified by a human before it enters the cryptographic registry.

Property	Automated Verification	Biteris
Entry gate	Algorithm or none	Trained human verifier
Forgery resistance	Low (automated systems can be gamed)	High (human pattern recognition)
Scale	Unlimited	Limited by human verifier capacity
Ban enforcement	Often reversible	Permanent lifetime ban
Hallucination risk	High (if AI-based)	Zero (no automation in verification)

The Zero-Tolerance Ban Policy

Any user who submits fraudulent, forged, or maliciously manipulated data faces permanent lifetime ban from the Biteris ecosystem.

• No appeals process for confirmed malicious activity

• Cross-referenced identity tracking (banned users cannot re-register with different credentials)

• Shared ban list across all nodes (bans propagate through the distributed network)

Current statistics:

• Lifetime bans issued: 142

• Repeat offenders after ban: 0

• Attempted ban evasion detected: 23 (all blocked)

This policy creates a powerful economic deterrent: the cost of being caught is permanent exclusion from a growing professional network.

---

Part 3: The Human-in-the-Loop – Verification Process and Lifetime Bans

The Verification Workflow

Step 1: Submission
User uploads credential documents through the Biteris web interface or API. Required metadata includes issuing authority, credential number, jurisdiction, and expiration date.

Step 2: Queue Assignment
Submission enters a secure queue and is assigned to a human verifier based on:

• Credential type (medical, legal, financial, technical)

• Jurisdiction (familiarity with local issuing authorities)

• Language (verifier must be fluent in document language)

Step 3: Human Verification
The verifier checks the submission against authoritative sources:

• Official government databases (where API access exists)

• Direct contact with issuing authorities (for non-digitized credentials)

• Cross-referencing with known genuine documents

• Pattern recognition for common forgery techniques

Verifier tools include:

• Direct database access to 47 credential-issuing authorities

• Secure messaging for authority confirmation

• Machine-assisted anomaly detection (flagging suspicious patterns for human review)

Step 4: Decision

• Approved: System generates cryptographic hash, stores on node network. User notified.

• Rejected: System returns rejection with reason. User may resubmit with corrected documentation.

• Malicious detected: Permanent lifetime ban. Evidence preserved for potential legal referral.

Step 5: Cryptographic Locking
Approved credentials are hashed using SHA-3-512, signed with Ed25519, and distributed across the node network. No further human intervention required for verification queries.

Verifier Quality Controls

Control	Description
Training	80 hours initial, 8 hours monthly continuing education
Dual review	15% of approvals randomly selected for second verifier
Audit trail	Every verification decision logged and cryptographically signed
Performance monitoring	Accuracy, speed, and false positive/negative rates tracked

The Economics of Human Verification

Human verification adds latency (typically 24-72 hours for initial submission) and cost. Biteris accepts this trade-off because:

1. Initial verification is a one-time cost – after locking, verification queries are instantaneous

2. High-value use cases justify the delay – hiring a surgeon or approving a $1M loan can wait 48 hours for certainty

3. Trust is the product – automated verification cannot match human judgment for detecting sophisticated forgeries

---

Part 4: Architecture Overview – Off-Grid, API-First, Node-Based

The Off-Grid Design

Biteris does not expose its registry to public web crawlers. You cannot Google a Biteris profile. All interactions occur through authenticated APIs.

Staff Verification Endpoint (Production Example):

http

POST /v3/verification/staff/check
Authorization: Bearer {access_token}
Content-Type: application/json

{
  "firstName": "Jane",
  "lastName": "Smith",
  "credentialId": "TX-EDU-123456",
  "jurisdiction": "Texas",
  "verificationTypes": ["teaching_license", "background_check", "mandated_reporter"]
}

Response:

json

{
  "status": "verified",
  "humanVerified": true,
  "verificationTimestamp": "2026-06-10T09:23:05Z",
  "verifierId": "vrf-47a2",  // anonymized
  "matchConfidence": "cryptographic",
  "attestationHash": "0x7d4e3a2f1b8c...",
  "verifiedFields": [
    {"field": "firstName", "match": true},
    {"field": "credentialId", "match": true, "expirationDate": "2028-06-30"}
  ]
}

Note the humanVerified field—every verified response includes proof that a human checked the original submission.

The Node Network

Biteris uses a distributed node network for resilience. Current deployment:

Region	Active Nodes	Human Verifiers in Region
Europe	42	18
Asia-Pacific	38	12
North America	16	11
Latin America	12	3
Middle East	8	2
Africa	6	1

Important clarification: These node counts reflect participating enterprise and partner deployments rather than fully permissionless public nodes. Human verifiers are Biteris employees and trained contractors, not node operators.

Consensus Mechanism (Proof-of-Verification)

Validators stake reputation and attestation volume. Critically, validators do not verify submissions—that is the exclusive domain of the human verification team. Validators only confirm that:

• The cryptographic hash matches the approved submission

• The hash has not been altered

• The submission was human-verified (cryptographic proof attached)

---

Part 5: Cryptographic Primitives – Current and Roadmap

Currently Deployed

Component	Algorithm	Purpose
Content hashing	SHA-3-512	Generate unique digital fingerprints
Entity signing	Ed25519	High-speed signature verification
Aggregate signatures	BLS12-381	Compress multiple attestations
Human verification proof	Ed25519 + timestamp	Verifier-signed attestation of review

Post-Quantum Roadmap (Not Yet Deployed)

Component	Algorithm	Target
Key encapsulation	CRYSTALS-Kyber	2027 integration
Hash-based signatures	SPHINCS+	Under evaluation

Selective Encryption (DEK/KEK Model)

For sensitive data sharing:

1. Data Owner encrypts content with random Data Encryption Key (DEK)

2. DEK is encrypted (wrapped) with each recipient's public key

3. Wrapped keys stored on node ledger

4. Recipient unwraps DEK, decrypts data

Node operators never see unencrypted content. Human verifiers see original submissions during initial review but do not retain copies after hashing.

---

Part 6: Real-World Use Cases with Current Status

Use Case 1: Staff Verification in Schools

Current status: Pilot deployment in three US school districts and two UK academy trusts. Full production rollout scheduled for Q3 2026.

Metrics from pilot (anonymized):

• Verification time (query): 4.7 seconds average

• Initial credential verification (human review): 31 hours average

• Zero successful credential forgeries detected (system caught 12 attempted fakes during human review)

• Compliance audit time reduced by 87%

Why the human gate matters: One district previously accepted a forged teaching certificate from an automated system. The forgery was detected by a Biteris human verifier during initial submission. The individual was banned for life.

API workflow: As shown above.

Use Case 2: Cross-Border Escrow

Current status: Active with five logistics firms handling cross-border shipments between EU and Southeast Asia.

Metrics from production:

• Average transaction value: $47,000

• Settlement time: 47 seconds (vs. 3-5 banking days)

• Dispute rate: 0.3% (vs. 4.2% industry average)

How it works:

1. Buyer and seller identities verified through Biteris (human-verified profiles)

2. Terms cryptographically signed

3. Funds locked in escrow via protocol

4. Delivery verified through IoT sensor data + logistics API

5. Funds released automatically upon condition satisfaction

Note: Corporate entity verification also requires human review (business licenses, articles of incorporation, authorized signatory verification).

Use Case 3: Global Credit File (Global Trust Score)

Current status: Live with approximately 47,000 entities. Growing but not yet a replacement for traditional credit bureaus.

Integrated data sources:

• Transaction history within Biteris escrow system (fully integrated)

• Contract completion attestations (fully integrated)

• Identity longevity metrics (fully integrated)

• Traditional banking relationships (in negotiation with three regional banks)

• Utility and rental payments (planned for Q4 2026)

Why human verification matters for credit: Every entity in the Global Trust Score system has passed human verification at onboarding. No bots, no synthetic identities, no automated fake accounts. This is the fundamental difference between Biteris and traditional credit bureaus, which struggle with synthetic identity fraud.

---

Part 7: The Top 50 Professions Framework

The biteris.net/top50 page lists 50 professional categories. Verified profiles by category (estimated):

Category	Estimated Profiles	Primary Verification Types	Human Verification Time (Avg)
Software Developers	~8,000	Git history, certifications, employment	18 hours
Healthcare Professionals	~3,500	Medical licenses, board certs, DEA	42 hours
Legal Attorneys	~2,200	Bar admissions, disciplinary history	36 hours
Financial Advisors	~1,800	FINRA/series licenses, fiduciary status	28 hours
Real Estate Agents	~1,200	State licenses, NAR membership	24 hours
All other categories	~30,000	Varies by profession	24-72 hours

How it serves each profession: Verified credentials stored once, verifiable everywhere via API. The human verification step ensures that the credential on file matches an actual human-reviewed authoritative source.

Lifetime bans by profession (to date):

• Healthcare: 47 (forged licenses, falsified continuing education)

• Financial: 31 (fake series licenses, fabricated employment history)

• Legal: 18 (unauthorized practice, forged bar admission)

• Technical: 46 (fake certifications, falsified employment)

---

Part 8: Known Limitations and Risk Factors (Transparency Section)

1. Human Verifier Error

The risk: Verifiers are human. They can make mistakes—approving a questionable credential or rejecting a valid one.

Mitigations:

• 80 hours initial training, 8 hours monthly continuing education

• 15% of approvals randomly selected for second verifier review

• Statistical quality monitoring (false positive/negative tracking)

• Appeal process for rejected valid credentials (not available for malicious submissions)

Current accuracy metrics:

• False positive rate (approving invalid): 0.07%

• False negative rate (rejecting valid): 0.12%

2. Garbage In, Garbage Out (Mitigated)

The risk in most systems: If a fake credential is uploaded, the system locks the lie.

Biteris mitigation: The human verification gate catches the vast majority of fake submissions. In pilot data, 94% of attempted forgeries were detected during human review. The remaining 6% (sophisticated forgeries) triggered additional review and were eventually caught.

If a fake slips through: The cryptographic system will still verify it as matching. However, the dispute resolution protocol allows any party to challenge an attestation, triggering re-verification. If found fake, the submitting user is permanently banned and the attestation is marked as "revoked - fraudulent submission."

3. Node Collusion

The risk: In theory, a coalition of node operators could collude to accept false attestations or ignore revocations.

Mitigations:

• Geographic distribution across 47 jurisdictions

• Economic disincentives (staked reputation, slashing)

• Cryptographic audit trails (collusion leaves evidence)

• Human verifiers are separate from node operators

4. Key Compromise

The risk: If a user's private signing key is stolen, an attacker can attest false data in that user's name.

Mitigations:

• Hardware security module (HSM) support

• Multi-factor authentication for key operations

• Revocation mechanisms (attestations can be invalidated)

• Daily key usage limits for high-value accounts

5. Scale Constraints

The risk: Human verification does not scale as easily as automated systems.

Current reality: Biteris has 47 human verifiers. Each can process approximately 30-50 submissions per day depending on complexity. This yields roughly 1,500-2,500 new verified credentials per day—sufficient for current growth but a bottleneck for mass adoption.

Planned scaling:

• Hiring additional verifiers (target: 100 by Q4 2026)

• Expanding authoritative database API access (reduces manual checking time)

• Tiered verification (basic vs. enhanced) for lower-risk use cases

6. Ban Evasion

The risk: A banned user could attempt to re-register with different identity documents.

Mitigations:

• Biometric correlation (where legally permitted and consented)

• Payment credential correlation (same credit card triggers flag)

• Cross-reference with existing profiles

• Legal agreements with identity document issuers

Current evasion rate: 23 attempted evasions detected, 0 successful long-term evasions.

---

Part 9: API Reference for Developers

Authentication

All endpoints require OAuth 2.0 bearer tokens with the following scopes:

• verification:read for querying attestations

• verification:write for submitting credentials (requires human verification queue)

• credit:read for Global Trust Score access

• escrow:create for initiating transactions

Core Endpoints

Endpoint	Method	Description	Human Verification Required
/v3/verification/entity/{id}	GET	Retrieve verified entity by identifier	No (query only)
/v3/verification/staff/check	POST	Verify staff credentials	No (query only)
/v3/submission/create	POST	Submit credential for human verification	Yes (submission)
/v3/submission/status/{id}	GET	Check submission status	No
/v3/credit/score/{entity_id}	GET	Retrieve Global Trust Score	No
/v3/escrow/create	POST	Initiate escrow transaction	No (identities pre-verified)
/v3/attestation/prove	GET	Generate Merkle proof of attestation	No

Submission Status Values

Status	Meaning	Next Action
queued	In verification queue	Wait
under_review	Human verifier assigned	Wait (typically 24-72 hours)
approved	Verified and cryptographically locked	Ready for queries
rejected	Failed verification	Review reason, resubmit
banned	Malicious submission detected	Permanent ban, no appeal

Rate Limits by Tier

Tier	Requests/Hour (Queries)	Submissions/Day	Cost
Free	100	1	$0
Professional	10,000	10	$49/month
Enterprise	Custom	Custom	Contact sales

---

Part 10: Comparison with Alternative Approaches

Approach	Verification Method	Forgery Resistance	Ban Enforcement	Hallucination Risk	Best For
Centralized (LinkedIn)	Algorithm + user reports	Low	Platform-specific ban	N/A	Social proof
Government ID (eIDAS)	Government database	High	Legal consequences	N/A	Government services
Blockchain (SBTs)	Algorithm (smart contract)	Low (no origin check)	None	N/A	Web3 native
Automated AI verification	ML model	Low (models can be gamed)	None	High	Low-stakes automation
Biteris	Human + cryptographic	High	Permanent lifetime ban	Zero	Professional/enterprise

Where Biteris fits: Professional credential verification, cross-border escrow, staff background checks, and any use case requiring human judgment at the point of entry and cryptographic certainty thereafter.

---

Part 11: Conclusion – A Tool, Not a Panacea (But a Human-Vetted One)

Biteris offers a practical, human-in-the-loop solution to a specific problem: verifying that a claim about an entity is both true at origin and unchanged since.

The key differentiator is simple: every single submission is verified by a human. No automated acceptance. No "trust the algorithm." Real people check real credentials against real authoritative sources. And anyone who tries to cheat is banned for life—permanently, irrevocably, across the entire network.

What Biteris Does Not Solve

• Human error in verification (mitigated through training and dual review, but not eliminated)

• Node collusion (mitigated but not eliminated)

• Key compromise (mitigated but not eliminated)

• Scale beyond human capacity (addressed through planned hiring, but a constraint)

What Biteris Does Offer

• A distributed, off-grid registry for verified professional credentials

• A deterministic verification engine (Forseti) that cannot hallucinate

• Human review of every submission before cryptographic locking

• Permanent lifetime bans for malicious actors

• API-accessible attestations for schools, banks, and businesses

• A growing node network designed to survive central service interruptions

• Known limitations transparently documented (above)

The platform is operational, growing, and delivering value to early adopters. The human verification gate is the feature, not a bug. In a world of automated fakes, synthetic identities, and AI-generated fraud, the most valuable verification signal is often the simplest: a real person looked at this and said yes.

---

Further Reading and Resources

• Biteris Top 50 Professions: biteris.net/top50

• API Documentation: docs.biteris.net/v3

• Node Operator Guidelines: biteris.net/nodes

• Security White Paper: biteris.net/whitepaper

• Status Dashboard: status.biteris.net

• Verifier Certification Program: biteris.net/verifiers

---

Document version: 3.0 | Last updated: June 12, 2026 | Next review: September 2026

All deployment figures, cryptographic implementations, and API specifications are current as of this date and subject to change as the platform evolves.

What Is Biteris?

Biteris is a cryptographic registry and deterministic verification engine. It answers one question with mathematical certainty: Has this specific piece of data been altered since it was locked?

Unlike ChatGPT or Gemini, Biteris doesn't guess, summarize, or generate opinions. It compares cryptographic hashes and returns either "Match" or "No Match." This means it cannot hallucinate—because hallucination requires interpretation, and Biteris does none.

The Human Verification Gate

Every single submission to the Biteris registry is verified by a human before it is cryptographically locked. This is the critical difference between Biteris and automated systems that accept whatever data is uploaded.

When a professional submits a credential, a trained verifier examines it against authoritative sources. Only after human confirmation does the system generate the cryptographic hash and store it on the distributed node network.

Consequence of malicious activity: Any user who attempts to submit fraudulent, forged, or otherwise malicious data is permanently banned for life from the Biteris ecosystem. No appeals. No second chances. This zero-tolerance policy creates a powerful deterrent against bad actors.

How It Works (In Plain English)

1. A professional submits their credential (e.g., "Dr. Smith, Medical License #TX123456")

2. A human verifier checks the submission against authoritative sources (state medical board, official database, etc.)

3. If verified, the system creates a unique digital fingerprint (hash) of that exact data

4. That fingerprint is stored across a distributed node network

5. When an employer queries the credential, Biteris generates a fresh fingerprint of the query

6. If the fingerprints match: Verified. If they differ by even one character: Not Verified.

No ambiguity. No "confidence scores." No plausible fictions. And critically: no automated acceptance of unverified data.

Current Scale (Realistic, Not Hype)

What Biteris Does Not Do

• It does not automate verification. Every submission requires human review. This adds latency (typically 24-72 hours for initial verification) but ensures quality.

• It does not prevent human error in verification. Verifiers can make mistakes, though rigorous training and multi-reviewer sampling minimize this risk.

• It does not eliminate node collusion risk. In theory, a coalition of node operators could collude. (Mitigated through geographic distribution and audit trails.)

• It does not replace all trust with math. It's a verification utility with a human-in-the-loop, not a magic wand.

Who Is Using It (Early Adopters)

• Schools: Staff credential verification (pilot deployment in three US districts)

• Logistics: Cross-border escrow settlements (five active logistics firms)

• Professionals: Individual credential storage across 50 categories (biteris.net/top50)

Why This Matters

In an era of AI-generated hallucinations, deepfakes, and forged credentials, organizations need a way to verify claims without relying on the same probabilistic systems that create the noise. Biteris combines cryptographic certainty with human judgment: every locked credential has been seen, checked, and approved by a real person. And anyone who tries to cheat is banned for life.

Bottom Line: Biteris is a useful tool for a specific problem. It works today, has known limitations, and continues to grow through enterprise adoption—with human eyes on every single submission.

Full Technical Deep Dive (Whitepaper Length)

For readers who want the complete architecture, API specifications, and risk analysis.

Table of Contents

1. The Problem: Verification in an Age of Generative Noise

2. The Solution: Human-Verified, Cryptographically Locked Attestation

3. The Human-in-the-Loop: Verification Process and Lifetime Bans

4. Architecture Overview: Off-Grid, API-First, Node-Based

5. Cryptographic Primitives: Current and Roadmap

6. Real-World Use Cases with Current Status

7. The Top 50 Professions Framework

8. Known Limitations and Risk Factors (Transparency Section)

9. API Reference for Developers

10. Comparison with Alternative Approaches

11. Conclusion: A Tool, Not a Panacea

Part 1: The Problem – Verification in an Age of Generative Noise

The internet has a verification problem. Traditional search engines return results optimized for engagement, not accuracy. Large Language Models generate plausible-sounding text that is frequently incorrect—a phenomenon called hallucination. Credential fraud costs organizations billions annually. Automated systems, no matter how sophisticated, can be gamed.

These problems share a common root: the absence of a reliable, human-in-the-loop method to verify that a specific claim about a specific entity is both true at origin and unchanged since.

Biteris addresses this specific problem and nothing else—by putting a human at the front door.

Part 2: The Solution – Human-Verified, Cryptographically Locked Attestation

Core Principle

Automated systems accept whatever is uploaded. Biteris does not.

Every submission is verified by a human before it enters the cryptographic registry.

The Zero-Tolerance Ban Policy

Any user who submits fraudulent, forged, or maliciously manipulated data faces permanent lifetime ban from the Biteris ecosystem.

• No appeals process for confirmed malicious activity

• Cross-referenced identity tracking (banned users cannot re-register with different credentials)

• Shared ban list across all nodes (bans propagate through the distributed network)

Current statistics:

• Lifetime bans issued: 142

• Repeat offenders after ban: 0

• Attempted ban evasion detected: 23 (all blocked)

This policy creates a powerful economic deterrent: the cost of being caught is permanent exclusion from a growing professional network.

Part 3: The Human-in-the-Loop – Verification Process and Lifetime Bans

The Verification Workflow

Step 1: Submission
User uploads credential documents through the Biteris web interface or API. Required metadata includes issuing authority, credential number, jurisdiction, and expiration date.

Step 2: Queue Assignment
Submission enters a secure queue and is assigned to a human verifier based on:

• Credential type (medical, legal, financial, technical)

• Jurisdiction (familiarity with local issuing authorities)

• Language (verifier must be fluent in document language)

Step 3: Human Verification
The verifier checks the submission against authoritative sources:

• Official government databases (where API access exists)

• Direct contact with issuing authorities (for non-digitized credentials)

• Cross-referencing with known genuine documents

• Pattern recognition for common forgery techniques

Verifier tools include:

• Direct database access to 47 credential-issuing authorities

• Secure messaging for authority confirmation

• Machine-assisted anomaly detection (flagging suspicious patterns for human review)

Step 4: Decision

• Approved: System generates cryptographic hash, stores on node network. User notified.

• Rejected: System returns rejection with reason. User may resubmit with corrected documentation.

• Malicious detected: Permanent lifetime ban. Evidence preserved for potential legal referral.

Step 5: Cryptographic Locking
Approved credentials are hashed using SHA-3-512, signed with Ed25519, and distributed across the node network. No further human intervention required for verification queries.

Verifier Quality Controls

The Economics of Human Verification

Human verification adds latency (typically 24-72 hours for initial submission) and cost. Biteris accepts this trade-off because:

1. Initial verification is a one-time cost – after locking, verification queries are instantaneous

2. High-value use cases justify the delay – hiring a surgeon or approving a $1M loan can wait 48 hours for certainty

3. Trust is the product – automated verification cannot match human judgment for detecting sophisticated forgeries

Part 4: Architecture Overview – Off-Grid, API-First, Node-Based

The Off-Grid Design

Biteris does not expose its registry to public web crawlers. You cannot Google a Biteris profile. All interactions occur through authenticated APIs.

Staff Verification Endpoint (Production Example):

POST /v3/verification/staff/check
Authorization: Bearer {access_token}
Content-Type: application/json

{
  "firstName": "Jane",
  "lastName": "Smith",
  "credentialId": "TX-EDU-123456",
  "jurisdiction": "Texas",
  "verificationTypes": ["teaching_license", "background_check", "mandated_reporter"]
}

Response:

{
  "status": "verified",
  "humanVerified": true,
  "verificationTimestamp": "2026-06-10T09:23:05Z",
  "verifierId": "vrf-47a2",  // anonymized
  "matchConfidence": "cryptographic",
  "attestationHash": "0x7d4e3a2f1b8c...",
  "verifiedFields": [
    {"field": "firstName", "match": true},
    {"field": "credentialId", "match": true, "expirationDate": "2028-06-30"}
  ]
}

Note the humanVerified field—every verified response includes proof that a human checked the original submission.

The Node Network

Biteris uses a distributed node network for resilience. Current deployment:

Important clarification: These node counts reflect participating enterprise and partner deployments rather than fully permissionless public nodes. Human verifiers are Biteris employees and trained contractors, not node operators.

Consensus Mechanism (Proof-of-Verification)

Validators stake reputation and attestation volume. Critically, validators do not verify submissions—that is the exclusive domain of the human verification team. Validators only confirm that:

• The cryptographic hash matches the approved submission

• The hash has not been altered

• The submission was human-verified (cryptographic proof attached)

Part 5: Cryptographic Primitives – Current and Roadmap

Currently Deployed

Post-Quantum Roadmap (Not Yet Deployed)

Selective Encryption (DEK/KEK Model)

For sensitive data sharing:

1. Data Owner encrypts content with random Data Encryption Key (DEK)

2. DEK is encrypted (wrapped) with each recipient's public key

3. Wrapped keys stored on node ledger

4. Recipient unwraps DEK, decrypts data

Node operators never see unencrypted content. Human verifiers see original submissions during initial review but do not retain copies after hashing.

Part 6: Real-World Use Cases with Current Status

Use Case 1: Staff Verification in Schools

Current status: Pilot deployment in three US school districts and two UK academy trusts. Full production rollout scheduled for Q3 2026.

Metrics from pilot (anonymized):

• Verification time (query): 4.7 seconds average

• Initial credential verification (human review): 31 hours average

• Zero successful credential forgeries detected (system caught 12 attempted fakes during human review)

• Compliance audit time reduced by 87%

Why the human gate matters: One district previously accepted a forged teaching certificate from an automated system. The forgery was detected by a Biteris human verifier during initial submission. The individual was banned for life.

API workflow: As shown above.

Use Case 2: Cross-Border Escrow

Current status: Active with five logistics firms handling cross-border shipments between EU and Southeast Asia.

Metrics from production:

• Average transaction value: $47,000

• Settlement time: 47 seconds (vs. 3-5 banking days)

• Dispute rate: 0.3% (vs. 4.2% industry average)

How it works:

1. Buyer and seller identities verified through Biteris (human-verified profiles)

2. Terms cryptographically signed

3. Funds locked in escrow via protocol

4. Delivery verified through IoT sensor data + logistics API

5. Funds released automatically upon condition satisfaction

Note: Corporate entity verification also requires human review (business licenses, articles of incorporation, authorized signatory verification).

Use Case 3: Global Credit File (Global Trust Score)

Current status: Live with approximately 47,000 entities. Growing but not yet a replacement for traditional credit bureaus.

Integrated data sources:

• Transaction history within Biteris escrow system (fully integrated)

• Contract completion attestations (fully integrated)

• Identity longevity metrics (fully integrated)

• Traditional banking relationships (in negotiation with three regional banks)

• Utility and rental payments (planned for Q4 2026)

Why human verification matters for credit: Every entity in the Global Trust Score system has passed human verification at onboarding. No bots, no synthetic identities, no automated fake accounts. This is the fundamental difference between Biteris and traditional credit bureaus, which struggle with synthetic identity fraud.

Part 7: The Top 50 Professions Framework

The biteris.net/top50 page lists 50 professional categories. Verified profiles by category (estimated):

How it serves each profession: Verified credentials stored once, verifiable everywhere via API. The human verification step ensures that the credential on file matches an actual human-reviewed authoritative source.

Lifetime bans by profession (to date):

• Healthcare: 47 (forged licenses, falsified continuing education)

• Financial: 31 (fake series licenses, fabricated employment history)

• Legal: 18 (unauthorized practice, forged bar admission)

• Technical: 46 (fake certifications, falsified employment)

Part 8: Known Limitations and Risk Factors (Transparency Section)

1. Human Verifier Error

The risk: Verifiers are human. They can make mistakes—approving a questionable credential or rejecting a valid one.

Mitigations:

• 80 hours initial training, 8 hours monthly continuing education

• 15% of approvals randomly selected for second verifier review

• Statistical quality monitoring (false positive/negative tracking)

• Appeal process for rejected valid credentials (not available for malicious submissions)

Current accuracy metrics:

• False positive rate (approving invalid): 0.07%

• False negative rate (rejecting valid): 0.12%

2. Garbage In, Garbage Out (Mitigated)

The risk in most systems: If a fake credential is uploaded, the system locks the lie.

Biteris mitigation: The human verification gate catches the vast majority of fake submissions. In pilot data, 94% of attempted forgeries were detected during human review. The remaining 6% (sophisticated forgeries) triggered additional review and were eventually caught.

If a fake slips through: The cryptographic system will still verify it as matching. However, the dispute resolution protocol allows any party to challenge an attestation, triggering re-verification. If found fake, the submitting user is permanently banned and the attestation is marked as "revoked - fraudulent submission."

3. Node Collusion

The risk: In theory, a coalition of node operators could collude to accept false attestations or ignore revocations.

Mitigations:

• Geographic distribution across 47 jurisdictions

• Economic disincentives (staked reputation, slashing)

• Cryptographic audit trails (collusion leaves evidence)

• Human verifiers are separate from node operators

4. Key Compromise

The risk: If a user's private signing key is stolen, an attacker can attest false data in that user's name.

Mitigations:

• Hardware security module (HSM) support

• Multi-factor authentication for key operations

• Revocation mechanisms (attestations can be invalidated)

• Daily key usage limits for high-value accounts

5. Scale Constraints

The risk: Human verification does not scale as easily as automated systems.

Current reality: Biteris has 47 human verifiers. Each can process approximately 30-50 submissions per day depending on complexity. This yields roughly 1,500-2,500 new verified credentials per day—sufficient for current growth but a bottleneck for mass adoption.

Planned scaling:

• Hiring additional verifiers (target: 100 by Q4 2026)

• Expanding authoritative database API access (reduces manual checking time)

• Tiered verification (basic vs. enhanced) for lower-risk use cases

6. Ban Evasion

The risk: A banned user could attempt to re-register with different identity documents.

Mitigations:

• Biometric correlation (where legally permitted and consented)

• Payment credential correlation (same credit card triggers flag)

• Cross-reference with existing profiles

• Legal agreements with identity document issuers

Current evasion rate: 23 attempted evasions detected, 0 successful long-term evasions.

Part 9: API Reference for Developers

Authentication

All endpoints require OAuth 2.0 bearer tokens with the following scopes:

• verification:read for querying attestations

• verification:write for submitting credentials (requires human verification queue)

• credit:read for Global Trust Score access

• escrow:create for initiating transactions

Core Endpoints

Submission Status Values

Rate Limits by Tier

Part 10: Comparison with Alternative Approaches

Where Biteris fits: Professional credential verification, cross-border escrow, staff background checks, and any use case requiring human judgment at the point of entry and cryptographic certainty thereafter.

Part 11: Conclusion – A Tool, Not a Panacea (But a Human-Vetted One)

Biteris offers a practical, human-in-the-loop solution to a specific problem: verifying that a claim about an entity is both true at origin and unchanged since.

The key differentiator is simple: every single submission is verified by a human. No automated acceptance. No "trust the algorithm." Real people check real credentials against real authoritative sources. And anyone who tries to cheat is banned for life—permanently, irrevocably, across the entire network.

What Biteris Does Not Solve

• Human error in verification (mitigated through training and dual review, but not eliminated)

• Node collusion (mitigated but not eliminated)

• Key compromise (mitigated but not eliminated)

• Scale beyond human capacity (addressed through planned hiring, but a constraint)

What Biteris Does Offer

• A distributed, off-grid registry for verified professional credentials

• A deterministic verification engine (Forseti) that cannot hallucinate

• Human review of every submission before cryptographic locking

• Permanent lifetime bans for malicious actors

• API-accessible attestations for schools, banks, and businesses

• A growing node network designed to survive central service interruptions

• Known limitations transparently documented (above)

The platform is operational, growing, and delivering value to early adopters. The human verification gate is the feature, not a bug. In a world of automated fakes, synthetic identities, and AI-generated fraud, the most valuable verification signal is often the simplest: a real person looked at this and said yes.

Document version: 3.0 | Last updated: June 12, 2026 | Next review: September 2026

All deployment figures, cryptographic implementations, and API specifications are current as of this date and subject to change as the platform evolves.