IamVERA

Detection

Detect potential AI hallucinations. Reduce sensitive-data exposure.

An answer can sound convincing while containing an unsupported claim, a missing qualification or a reference that does not support its conclusion.

Convincing

AI-generated answers can be fluent and plausible even when they are wrong. A convincing tone is not evidence that the content is correct.

Hidden

sensitive details may be present in the context sent to external models. The Semantic Privacy Shield replaces them before that happens.

"Do not judge an answer only by how confidently it is written. Examine how it was checked."

Scope

AI answer verification — what Vera is and what it isn't

Vera helps you examine AI-generated answers before you use them in your work. Its verification process checks the content; its Privacy Shield controls the release of transformed document context. Neither guarantees that every error or identifying detail will be detected.

What Vera is

An AI answer verification layer for professional work. Vera cross-checks one answer through independent models that audit facts, challenge reasoning and verify live sources, then shows every correction, disagreement and source behind the final answer.

What Vera is not

Not a single language model. Not a guarantee of truth. Not a replacement for professional judgement. Final responsibility for acting on an answer stays with you.

How it works

How multi-model AI verification works

Choose the checks that suit your task. With the full verification chain selected, each model has a distinct role.

01

Initial answer

Claude drafts the initial response

02

Factual review

GPT reviews factual claims for errors

03

Critical challenge

Grok challenges the reasoning and assumptions

04

Source verification

Perplexity checks relevant public sources

Claude creates the initial answer

Claude drafts an answer using your question and the context available for the task. This becomes the baseline that every subsequent model evaluates — not a final answer, but a starting point for structured verification.

Example output

The GDPR came into effect on 25 May 2018.

GPT performs a factual review

GPT (OpenAI) receives Claude's answer and reviews each factual claim for possible errors, inconsistencies and missing support. Corrections, confirmations and confidence levels are returned — not as a replacement answer, but as a structured review of the first response.

GPT fact-check output

✓ Confirmed: GDPR enforcement date is 25 May 2018.
Regulation (EU) 2016/679 — no errors detected in this claim.

Grok performs a critical challenge

Grok (xAI) examines assumptions, challenges the reasoning and raises alternative interpretations. It looks for missing context, ambiguous framing, and important distinctions that the first two models may have glossed over. Disagreement is a feature, not a bug.

Grok challenge output

⚠ Nuance: GDPR was adopted 27 Apr 2016 and entered into force 24 May 2016.
"25 May 2018" is the date of application. Entry into force ≠ application — the distinction matters in legal contexts.

Perplexity verifies sources

Perplexity checks relevant claims against available current public sources. Claude then incorporates the review findings into the final answer — with every correction and source visible.

Final verified answer

GDPR adopted 27 Apr 2016, entered into force 24 May 2016, applicable since 25 May 2018.
Source: EUR-Lex, Regulation (EU) 2016/679

Radical transparency

See the checks behind the answer

Vera records the selected model steps, review findings, corrections and sources. You can examine what was questioned, what changed and where assessments differed — rather than relying on the final text alone.

Q: When did the GDPR come into effect, and what regulation number is it?

Vera — Verified answer

The GDPR was adopted on 27 April 2016, entered into force on 24 May 2016, and became applicable on 25 May 2018. It is formally known as Regulation (EU) 2016/679. (Source: EUR-Lex)

C

Claude — Initial answer

The GDPR came into effect on 25 May 2018.

G

GPT — Fact-check

✓ Enforcement date confirmed. Regulation number: (EU) 2016/679 — added for completeness.

X

Grok — Challenge

⚠ Distinction flagged: GDPR adopted 27 Apr 2016, entered into force 24 May 2016, applicable 25 May 2018. Entry into force ≠ application. Relevant in legal contexts.

P

Perplexity — Source check

EUR-Lex confirms: adopted 27 Apr 2016, entered into force 24 May 2016, applicable since 25 May 2018. Grok's distinction verified.

Evidence

Inspect the evidence behind the product

We publish tests of our verification chain, including unsuccessful checks and the underlying run data.

These are tests of our own system, with small samples and results that can vary between runs and model versions. They show observed behaviour, not a guarantee that your next answer will be correct.

Method

Tests for factual traps and potential hallucinations

False-premise traps, source grounding, prompt-injection resistance and adversarial pressure to guess — each run through the configured chain and scored by deterministic checks.

Honesty

Failures and disagreements are visible

Where a check fails or the models disagree, we show it. A benchmark that only ever passes proves little; the disagreements are where the chain earns its keep.

Raw data

Inspect the raw verification data

Every result links to the underlying JSON: model calls, corrections, costs and judge output. Nothing summarised away.

Control

Choose the plan that fits your work

Starter is free for three months: 10 runs per month, no credit card required. Compare subscriptions, included verification runs and document-processing options on our pricing page. Review the current prices and allowances before choosing your plan.

C

Claude

Answer

G

GPT

Fact-check

X

Grok

Challenge

P

Perplexity

Source check

Claude (answer)~$0.003

GPT (fact-check)~$0.002

Grok (challenge)~$0.001

Perplexity (sources)~$0.005

This verification~$0.011

Semantic Privacy Shield

Confidential documents, reduced exposure

The Privacy Shield is a separate utility inside Vera. Before document context reaches an external model, it replaces detected sensitive values with synthetic placeholders. The AI chain receives the transformed context; the original values are restored locally. It is a privacy-control workflow, not an end-to-end guarantee against leakage.

1 · Detect and pseudonymise locally

Names, dates, references to minors, diagnoses, criminal-law context, case numbers — replaced locally with realistic synthetic values. Roles and relationships stay intact, so the AI still understands the file.

2 · Verify the outbound context

A local check confirms no detected real values remain in the prepared text. If it is not clean, nothing is sent. Fail closed — no exceptions.

3 · Restore original values locally

The verified answer is restored locally with the original values. You get a normal, usable answer — the public AI never saw the real data.

// Sent to the AI chain — fictional example
Mrs Dylan Knoers [[DOC001_ADULT_PERSON_001]] states that her ex-partner Mr Tobias Evers [[DOC001_ADULT_PERSON_002]] became aggressive on 06-01-2026 [[DOC001_DATE_001]] during the handover of her minor daughter [[DOC001_MINOR_CHILD_001]].

Built for professionals bound by confidentiality — lawyers, notaries, judges, physicians, occupational health physicians, psychiatrists, investigative journalists.

Vera X-Ray

Your confidential file never leaves your computer

The Privacy Shield protects documents inside Vera. Vera X-Ray starts one step earlier: on your own device. It detects sensitive values in a document, lets you review what will be shared, and sends Vera only the context you approved.

• Detect locally Names, addresses, account numbers, case references and supported secrets are found on your device, in PDFs, scanned pages and text.
• Review before release You see exactly what X-Ray found, mark what it missed and decide what goes out.
• Restore locally Vera answers with references. The real values are put back on your device, visible only to you.

PDF & OCR · Dutch & English · Human review before release

Processing

Clear about where processing happens

Your Vera environment and document-protection processing are hosted in the EU. Your instructions and approved, transformed context can reach selected external AI providers. Those providers may process information outside the EU. "Local" therefore means within your protected Vera environment, not necessarily on your own device.

Privacy & control

Designed for data control. No hidden flows.

A verification tool that obscures its own data handling would undermine the very thing it promises. So we don't.

Temporary chat PDFs

Chat-uploaded PDFs are processed temporarily in memory for the active run; metadata may be retained for session history.

Persistent Office documents

Vera Office documents remain in the customer-isolated environment until an authorised user deletes them.

Transparent model routing

Model calls — including document content in the context — are sent to the configured AI providers. This is stated clearly, not buried in a privacy policy.

Built in the EU

Vera is built and operated within the EU. We are working towards full GDPR processor documentation. We will only claim compliance when the documentation is complete.

Use cases

Built for professionals who cannot rely on one unchecked model

Vera is designed for work where AI output is acted upon — not just read.

Legal

Legal adviser

Have a draft analysis reviewed for unsupported claims, unclear wording and missing qualifications before you incorporate it into advice or an internal document.

Legal

Lawyer

Challenge the reasoning in a draft argument, examine cited public sources and identify questions that require further legal research before you use the result.

Business

Board member

Review a report or proposal for assumptions, missing information and unsupported conclusions before it becomes part of your decision-making.

Healthcare

Doctor

Structure a report or make an existing explanation easier to understand. Keep the task focused on documentation, with clinical judgement remaining yours.

Healthcare

Healthcare professional

Turn supplied notes into a clearer draft summary or handover. Review whether observations, interpretations, agreements and outstanding questions have been kept distinct.

Property

Estate agent

Review property descriptions, correspondence and agreement summaries for inconsistencies or unsupported statements before sharing them with your client.

Legal

Notary

Give a draft deed or explanatory note an additional review for inconsistent wording, unclear references and missing information before completing your own professional checks.

These are suggested applications, not guarantees of specialist accuracy. The medical and care examples concern text support and documentation, not diagnosis or treatment decisions. Vera does not replace professional research, formal checks or your own judgement.

Starter

Free

3 months free — then upgrade

• 10 verification runs per month
• 1 user
• room for roughly 400,000 credits per month
• No Vera Office
• No API
• No verification reports
• No audit trail
• No support
• No discount on extra runs
• No X-Ray support

Essential

€79.00 per month €65.83 per month

Pay yearly: 17% off €790.00 per year, paid once — 17% off (2 months free), you save €158.00 per year

• 50 verification runs per month
• €1.58 per run
• 1 user
• room for roughly 2 million credits per month
• Vera Office: draft and edit documents
• API available at extra cost: €50.00 per month

Professional

€199.00 per month €165.83 per month

Pay yearly: 17% off €1,990.00 per year, paid once — 17% off (2 months free), you save €398.00 per year

• 300 verification runs per month
• €0.66 per run
• 5 users
• extra user €19.95 per month, adding 100 runs
• room for roughly 12 million credits per month
• Data processing agreement on request
• Vera Office: draft and edit documents
• API available at extra cost: €50.00 per month
• 20% off extra runs

Compliance

€399.00 per month €332.50 per month

Pay yearly: 17% off €3,990.00 per year, paid once — 17% off (2 months free), you save €798.00 per year

• 1,000 verification runs per month
• €0.40 per run
• 25 users
• extra user €24.95 per month, adding 100 runs
• room for roughly 40 million credits per month
• 250 documents via the GPU server
• Data processing agreement included
• Priority support
• Vera Office: draft and edit documents
• API included
• Verification report as PDF, to hand over
• 40% off extra runs

All amounts exclude VAT. Cancel monthly. What counts as a run, and buying extra runs →

FAQ

AI answer verification: common questions

Direct answers about multi-model AI verification, potential hallucinations, source checking and confidential documents.

Certifications

Independently certified

Our processes are audited by independent parties against international standards.

ISO/IEC 27001

Information security: how we protect data, systems and access, with controls that are audited every year.

ISO 9001

Quality management: documented, repeatable processes with continuous improvement built in.

ISO/IEC 42001

AI management: responsible development and use of AI, including risk assessment and human oversight.

NEN 7510

Information security in Dutch healthcare: additional requirements for handling health data.