AI and automated processing disclosure

Where automation is used

• Intake: structuring your answers into a case profile and deciding which questions to ask next.
• Eligibility screening: producing an indicative route recommendation and a confidence score.
• Pricing: producing an indicative service-fee estimate and timeline range from reference data.
• Document handling: reading uploaded documents to identify the type, check for missing pages and flag expiry dates.
• Ava, our assistant: answering general questions about routes, requirements and pricing.
• Verification: a second automated pass that checks the first for inconsistencies and unsupported claims.

What automated output is, and is not

Automated output is informational assistance. It is not legal advice, not a determination of your eligibility, and not a prediction of what any authority will decide. Requirements change, government sources are sometimes ambiguous, and automated systems can produce confident wording for an incorrect conclusion. Always treat an automated answer as a starting point to be confirmed.

Human and professional review

No application is submitted on the strength of automated output alone. Where the confidence score is low, where a hard-stop condition is detected, such as prior refusals, overstays, criminal history or complex family circumstances, or where the matter requires regulated advice, the case is routed to a person, and to an independent licensed professional where the matter requires one. Final prices on matters flagged for review are confirmed by a person before you are asked to pay them.

No solely automated decisions with legal effect

Entry does not make decisions about your immigration status. Only the relevant authority does. Entry does not use solely automated processing to produce a decision with a legal or similarly significant effect on you. You may ask for human review of any automated assessment, ask how it was produced, contest it and correct the information behind it, by writing to concierge@entry.com.

Your data and AI providers

To produce these outputs, the relevant parts of your case data are processed by third-party model providers acting as our processors under contract. They are not permitted to use your data to train their models. Only the data needed for the task is sent, and identifying details are minimised where the task does not require them. The categories of recipients are listed in the privacy policy.

Accuracy and reporting problems

If an automated answer looks wrong, tell us. Reports are reviewed, and where a factual error is confirmed the underlying reference data is corrected. Entry's liability for automated output is governed by the terms of service; nothing here creates a warranty of accuracy.