ParliView: a research platform of University College Dublin (www.parliview.org and app.parliview.org)
Version: 4.0
Effective Date: 30 September 2026
Last Updated: 25 September 2026
Contents. 1 Who we are. 2 The Study. 3 Information we collect. 4 Why we use it and our legal bases. 5 Questions that may reveal political opinions. 6 How AI processes your information. 7 Who we share information with. 8 International transfers. 9 How long we keep information. 10 Security. 11 Withdrawing, and your rights. 12 Automated decisions and moderation. 13 Cookies and device storage. 14 People named in parliamentary records. 15 Children. 16 Changes to this notice. 17 If you are in the UK. 18 Complaints.
ParliView is provided by University College Dublin, National University of Ireland, Dublin, Belfield, Dublin 4, Ireland (“UCD”, “we”, “us”, “our”). UCD is the controller of the personal data described in this notice: it decides why and how the data is used and is responsible for protecting it. UCD is an Irish public university and a public authority under Irish data protection law.
ParliView is operated as part of UCD research study 025-HS-26-C-Cross, “ParliView User Testing and Evaluation”, approved by the UCD Human Research Ethics Committee and running to 31 December 2028. An amendment describing the Platform as launched was submitted to the Committee on 25 September 2026 and is under review. The Principal Investigator is Dr James Cross, School of Politics and International Relations.
Our project partners are the University of Strathclyde (Glasgow, United Kingdom) and Transparency International EU (Brussels, Belgium). The project is funded by a Google.org Impact Challenge grant administered by the Silicon Valley Community Foundation and awarded to the University of Strathclyde. Neither the partners nor the funder has editorial control over AI Output. The partners receive only anonymised and aggregated results, not personal data (section 7). The funder receives no personal data.
| Contact | What for |
|---|---|
| privacy@parliview.org | Data protection requests, withdrawal from the Study and privacy questions |
| UCD Data Protection Officer: gdpr@ucd.ie | Independent advice and concerns about how UCD handles personal data |
| corrections@parliview.org | Reports of inaccurate AI Output, including AI Output about a named person |
| support@parliview.org | General questions, complaints about ParliView and moderation appeals |
| accessibility@parliview.org | Accessibility feedback and requests for this notice in another format |
| Post | ParliView, School of Politics and International Relations, University College Dublin, Belfield, Dublin 4, D04 V1W8, Ireland |
Words with a special meaning. “ParliView” or “the Platform” means www.parliview.org and app.parliview.org. “The Study” means UCD study 025-HS-26-C-Cross. “You” and “participant” mean a person with a ParliView account or who has asked for one. An “Answer” is text ParliView generates in reply to your question. A “Summary” is AI-generated content about parliamentary business that is stored and shown to all participants, such as debate, procedure, event, adopted-text and topic summaries and suggested starter questions. “AI Output” means Answers and Summaries. “EP Material” means documents and data published by the European Parliament. “Parltrack Data” means data compiled by Parltrack (parltrack.org), mainly from European Parliament sources, used under the Open Database Licence. “Research datasets” means pseudonymised extracts of ParliView data used for analysis.
What the Study is for. The Study asks how people use an AI system to find and check information about the European Parliament. It looks at what people search for, how accurate and useful sourced Answers are, and how design choices (study conditions) affect accuracy and use. It also asks what this means for public access to parliamentary information. The Platform exists to conduct the Study. The ethics committee approved it as a study of real use over time, and the findings are published. Using ParliView is how you take part: there is no separate task, and no minimum amount of use.
Joining. Access is by approval. You ask for access through the request form at app.parliview.org. If your request is approved, you create an account with an email address and password, or by signing in with Google, and you are placed in a group. An event code or invitation link only places an approved participant in a group; it does not give access by itself.
Study conditions. As part of the Study, participants can be randomly assigned to additional groups beyond the default group. Groups may see different versions of ParliView: for example, a different AI model, different instructions to the model, or different settings. Assignment can depend on how you joined (for example, which event or invitation you used). We do not tell you which version you are using while the Study runs, because that could affect the results. Your group has no effect on your legal rights, on your access to the Platform or on anything outside ParliView. Your group assignment is recorded with your account and in our logs (section 3).
Taking part in the Study
Taking part is voluntary. Before you first use the Platform, and whenever we publish a new version, you are asked to confirm that you have read the Participant Terms and Information and this Privacy Notice. Confirming this is how you agree to take part in the Study, and we keep a record of it. That agreement is an ethical safeguard required by UCD’s research ethics rules. It is not the legal basis on which we process your personal data: that basis is explained in section 4, and it does not depend on your agreement. You can withdraw from the Study at any time. Section 11 explains how, and what happens to your data when you do.
| Category | What it includes | When |
|---|---|---|
| Access request | Email address, name and country (all required), and our decision. If we decline your request, we keep the request and our decision for 6 months so that we can recognise a repeat request, then delete it. | When you ask for access |
| Account details | Email address; the code we email you to confirm it; multi-factor authentication settings if you turn them on; any event code or invitation link you use. Your password is held by our sign-in system, Amazon Cognito, not in ParliView’s own databases | When you create and use your account |
| Questions and conversations | Your questions, the Answers you receive, conversation titles, the order of messages, timestamps, and changes you make, such as renaming a conversation or removing it from view | When you use ParliView |
| Feedback | Thumbs up or down; the reason you choose after a thumbs down and any improvement you suggest; comments; and messages sent through the general feedback form. We also keep a copy of the conversation as it was when you gave feedback. That copy stays if you later remove the conversation from view | When you give feedback |
| Surveys (optional) | Your answers and when you submitted them. If you choose to give an email address for future research contact, it is kept separately from your answers | If you take part in a survey |
| Messages to us | Your email address, what you write and our replies, including data protection requests, withdrawal requests and correction reports | When you contact us |
| Category | What it includes | Source |
|---|---|---|
| Session data | A random session identifier in a cookie (sid, section 13). On our servers the session holds your sign-in tokens and basic profile details, such as email address and name, for 5 days from your last activity | Your browser; our sign-in system |
| Application logs | User, session, request, group, sub-group and conversation identifiers; the text of your questions; the search queries ParliView derives from them; the addresses of pages and documents the application requests; the reports sent when you open a source document (see below); error details. Your email address may appear in some log entries (for example when you log in or request access) for the log retention period | ParliView |
| Traces | Technical records of how a request moved through our systems, including user, session, request and group identifiers. They may also include your IP address and browser details | ParliView |
| Network and security logs | Our firewall logs record the IP address, date and time, the address requested, browser details (user agent) and the firewall’s decision. Load balancer and website access logs, where enabled, record similar details | Your browser; Amazon Web Services |
ParliView does not record which pages of the Platform you view. It does not record when you copy text.
Source documents you open. When you open a European Parliament document from ParliView, the app sends a short report to our servers. The report contains the address (URL) of the document, the type of view and the time. It is logged with your pseudonymous user, session, group and conversation identifiers. This is Study data: it shows whether participants check sources. It is not needed to show you the document. The reports are kept for the application log period (section 9) and are included in research datasets. To object, email privacy@parliview.org. We then exclude your records from research extracts, and your access is not affected. The app still sends the reports, and they stay in the logs until the log period ends. To stop them, you can withdraw (section 11).
| Category | What it includes |
|---|---|
| User ID | A random identifier created by our sign-in system. We use it instead of your name or email address in conversations, feedback, usage records and research datasets, and in most log entries (section 3.2). Authorised staff can link it back to your account |
| Group assignment | The groups and sub-groups your account belongs to, when you were assigned, and the version of the Platform (model, instructions and settings) that answered you |
| Acceptance records | The version and digital fingerprint (hash) of the Participant Terms and Information and this notice that you confirmed, when you did so, and the session, request and group identifiers for that visit |
| Usage records | For each day: the number of requests you made, the processing capacity (tokens) used, errors, and the experiment version that applied |
| Answers | The text generated in reply to your questions, which may contain personal data if your question does |
| Moderation records | If an automated check flags a question (section 12): up to 300 characters of the question as written and after cleaning, the rule or keyword matched, and your user ID. We also record our reviewer’s decision and, if we restrict your account, the reason and any end date |
| Research datasets | Pseudonymised extracts of the records above, with your user ID replaced by a research code and email addresses removed. Pseudonymised records are still personal data |
| What Google sends us | Notes |
|---|---|
| Your email address and your name | We do not ask Google for your phone number. Google also learns that you signed in to ParliView; Google’s own privacy policy governs what Google does with that |
No law requires you to. You cannot use the Platform without an approved access request and an account, and you cannot take part in the Study without using the Platform. Surveys and future research contact are optional; saying no has no effect on your access.
We do not collect payment details, precise location, or voice, video or biometric data. We do not use third-party analytics, advertising or tracking tools.
We use your personal data for two purposes, both of which are research:
Public task. UCD is a public university. Under the Universities Act 1997, its functions include advancing knowledge through research and disseminating the outcomes of research (section 12), and it must promote and facilitate research (section 13(2)(b)). The Study carries out those functions. For the two purposes above, our legal basis is public task. That is Article 6(1)(e) of the General Data Protection Regulation (GDPR), with section 38(1)(a) of the Data Protection Act 2018. We use personal data only where it is necessary and proportionate for those functions. As a public authority, UCD does not rely on “legitimate interests” for its tasks.
We consider operating the Platform necessary for the Study because it measures real use over time, which cannot be observed in a laboratory setting. We consider it proportionate because access is limited to approved participants, data is pseudonymised for analysis and retention is limited (section 9).
Research safeguards. Because both purposes are research, we apply the safeguards that Article 89(1) GDPR and section 42 of the 2018 Act require:
Your agreement to take part is not the legal basis. Confirming the participant information at the acceptance gate is required by research ethics and is one of our safeguards. It is not consent under Article 6(1)(a) GDPR, and our legal basis does not depend on it. Consent is our legal basis only for the optional items in the table below.
| Purpose | Information used | Legal basis (EU and EEA) |
|---|---|---|
| Handle your access request and the allowlist | Access request | Public task |
| Create and run your account: sign-in, email confirmation, multi-factor authentication, group access, welcome email | Account details; information from Google; session data | Public task |
| Answer your questions, keep your conversations and show Summaries | Questions and conversations; group assignment; usage records | Public task. For questions that may reveal political opinions, see section 5 |
| Assign you to study conditions and record which version answered you | Group assignment; usage records | Public task |
| Record your confirmation of the Participant Terms and Information and this notice | Acceptance records | Public task |
| Keep the Platform and your account secure, prevent abuse and apply usage limits | Session data; application logs; traces; network and security logs; usage records | Public task |
| Moderate questions, review any automated flags, restrict accounts and handle appeals | Moderation records; the questions concerned; account details | Public task (section 12) |
| Correct inaccurate AI Output and keep a record of corrections | Correction reports; the AI Output concerned | Public task |
| Study data on whether participants check sources | The reports sent when you open a source document (section 3.2) | Public task, with the research safeguards above. Section 13 explains the separate rule on information sent from your device |
| Analyse Study data: usage, questions, Answers and feedback, in pseudonymised form | Research datasets | Public task, with the research safeguards above. For political opinions, see section 5 |
| Prepare anonymised and aggregated results to share with our partners; publish aggregated findings; deposit anonymised datasets | Research datasets | Public task, with the research safeguards above |
| Reply to your messages, complaints and correction reports | Messages to us; the records concerned | Public task |
| Handle your data protection requests and meet other legal obligations, such as a regulator’s request or breach reporting | The records concerned | Legal obligation (Article 6(1)(c) GDPR) |
| Run optional surveys | Survey answers | Your consent (Article 6(1)(a) GDPR). If a survey asks about political views, your explicit consent (Article 9(2)(a) GDPR) |
| Contact you about future research | Future research contact email | Your consent (Article 6(1)(a) GDPR) |
| Establish, exercise or defend legal claims | The records concerned | Public task; for special category data, Article 9(2)(f) GDPR |
If we plan to use your information for a new purpose that this notice does not describe, we will tell you first.
Your right to object. Because we rely on public task, you can object at any time on grounds relating to your particular situation. We assess each objection on its own facts. We will stop unless we can show compelling legitimate grounds that override your interests, rights and freedoms, or we need the data for legal claims. If you object to the processing we need to run the Platform, we cannot keep your account open and will close it (section 11). If your objection is limited to research use of data we have already collected, we exclude your records from future research datasets and your account can stay open. To object, email privacy@parliview.org.
Your questions may reveal your political opinions, for example if you ask why an MEP voted a particular way or say what you think of a policy. Information that reveals political opinions is a special category of personal data under Article 9 GDPR and needs extra protection. This applies whatever you intended, and whether or not a conclusion drawn from your question would be right.
Legal basis. The GDPR allows special category data to be processed for scientific research where national law permits it and safeguards are in place. In Ireland, section 54 of the Data Protection Act 2018 permits this where the processing is necessary and proportionate for research, subject to the safeguards in section 42. Two steps involve this data:
We do not rely on your consent for either step, and you can object to either (section 4).
How we limit the impact. In addition to the research safeguards in section 4:
What to leave out. ParliView is for questions about the European Parliament. Please do not include personal details about yourself, such as your health, or information about other private individuals, in your questions. Answers and Summaries are about public parliamentary business, and ParliView is not designed to hold personal information about you or the people you know. If you include such details by mistake, you can remove the conversation from view, and you can ask us to delete it (section 11).
When you ask a question, ParliView:
What the models do not receive. Your name, email address, user ID and IP address are not sent to the models. The model and the instructions it is given can differ between study groups (section 2).
Where. Amazon Web Services (AWS) runs the models for us as our processor (section 7). Live Answers use Amazon Bedrock’s EU cross-region inference, which processes requests only in AWS data centres in the EU: Frankfurt, Stockholm, Milan, Spain, Ireland and Paris. Embeddings are created in Ireland.
Training and access. AWS states that the companies that make the models, Anthropic and Cohere, cannot access the questions sent to the models or the text they return, and that this content is not used to train models. We do not use models for which AWS keeps requests for abuse monitoring or human review.
Summaries are generated from EP Material and Parltrack Data when a participant first requests an item. They are then stored and shown to every participant who opens the same item. Summaries are produced in the same way for every participant, whatever study group they are in. They do not use participants’ questions or any other participant data. They are not reviewed by a person before they are shown. Summaries are generated through Amazon Bedrock in the EU, in the same way as Answers. When a Summary is corrected or withdrawn, we record the date and the reason.
Labels and limits. Every Answer and every Summary carries a label saying it is AI-generated. AI Output can be wrong, including about people. Please report errors to corrections@parliview.org. We do not use AI to make decisions about you (section 12). ParliView does not process voice, video or biometric data. Our AI Transparency and Accuracy page explains how the AI works, its limits and whether the generated text carries a machine-readable marking.
We share personal data only with the recipients below, and only as described.
| Recipient | Role | What they receive and why | Where | Safeguards |
|---|---|---|---|---|
| Amazon Web Services EMEA SARL, Luxembourg | Processor: acts only on our instructions | All the information in section 3, to host and run the Platform. The services are listed below this table | Ireland; EU data centres for AI processing (section 6); the global services named in section 8 | AWS Data Processing Addendum under Article 28 GDPR, including Standard Contractual Clauses |
| Google Ireland Limited | Independent controller | If you choose to sign in with Google, Google confirms your identity and sends us the details in section 3.4. Google’s use of your data is governed by the Google Privacy Policy | Google’s own terms | Not applicable: Google acts for itself |
| Public research repositories (for example Zenodo, operated by CERN in Switzerland) | Not applicable | Anonymised datasets only, prepared under a documented anonymisation process. Anonymised data is not personal data | Various | Not applicable |
| European Parliament | Independent controller (an EU institution with its own data protection rules) | Your IP address and browser details when your browser loads an MEP photo from the European Parliament’s servers (section 13), and any data you send when you follow a link to its websites | EU | The European Parliament’s own data protection notice |
| Proton AG (Proton Mail), Geneva, Switzerland | Processor | Emails you send to our @parliview.org addresses and our replies | Switzerland | The European Commission has decided that Switzerland gives adequate protection (Decision 2000/518/EC); Proton’s data processing agreement. |
| Courts, regulators, law enforcement bodies and UCD’s legal advisers | Independent controllers | Information we are legally required to disclose, or need to disclose to establish, exercise or defend legal claims | Usually Ireland | The law that applies to the request |
AWS services we use: sign-in (Amazon Cognito); databases (DynamoDB, Aurora); storage and backups (S3, AWS Backup); search (OpenSearch); email delivery (Amazon SES); logs and traces (CloudWatch, Firehose, X-Ray); computing (ECS, Lambda); load balancing and firewall; domain name service (Route 53); secrets management; AI models (Amazon Bedrock); hosting of www.parliview.org (S3, CloudFront).
We share only anonymised and aggregated results with our project partners, the University of Strathclyde and Transparency International EU. They do not receive participants’ personal data.
Anthropic and Cohere, which make the AI models, do not receive your data: the models run inside AWS (section 6). We do not sell personal data, share it for advertising, or show your conversations to other participants.
Our databases, logs and backups are held by AWS in Ireland, and AI processing of participant data takes place in AWS data centres in the EU (section 6). We do not rely on the EU-US Data Privacy Framework for any transfer.
| Transfer | What | Safeguard |
|---|---|---|
| To AWS global services, which process some data in the United States: content delivery for www.parliview.org (CloudFront) and AWS account activity logs (CloudTrail) | Technical data such as IP addresses, request details and administrator activity records. Not questions, conversations or account details | Standard Contractual Clauses adopted by the European Commission (Implementing Decision (EU) 2021/914), included in the AWS Data Processing Addendum (Article 46(2)(c) GDPR) |
Emails you send to our @parliview.org addresses, and our replies, are stored by Proton AG in Switzerland. The European Commission has decided that Switzerland gives adequate protection, so no further safeguard is needed. We make no other transfers of personal data outside the European Economic Area. If you sign in with Google, you give your details to Google directly and Google acts for itself. You can ask for a copy of the safeguards for any transfer by emailing privacy@parliview.org.
We keep personal data only as long as we need it for the purposes in section 4. When a period ends, we delete the data or anonymise it. “Study end” means 31 December 2028. “Withdrawal” means you asking us by email to close your account (section 11). We do not close accounts for inactivity. Where deletion is done by hand (access requests, moderation records, correction records and our correspondence), we delete the data in a quarterly review after its period ends.
| Information | How long we keep it | Why |
|---|---|---|
| Account (sign-in system), allowlist entry, group assignment | Until you withdraw by email, or the end of the Study on 31 December 2028; then deleted | Identifiers are not needed after the Study |
| Conversations, feedback, request history and usage records (live systems) | Until you withdraw by email, or the end of the Study on 31 December 2028. Pseudonymised research datasets extracted from them: until 31 December 2033 | The ethics committee approved keeping research data for five years after the Study so that findings can be checked; reviewed annually |
| Acceptance records | Study end plus 6 years. They are not deleted if you ask for erasure. Once your account is deleted they no longer link to you | Evidence of the information you were given |
| Moderation records (flags, restrictions) | 12 months after the decision | Abuse prevention |
| Access request (email, name, country, our decision) | 6 months after our decision, whether approved or declined. For declined requests, the email address is kept for that period so that a repeat request is recognised | Not needed after the decision |
| Application logs, including question text, identifiers and the reports sent when you open a source document (section 3.2) | Up to 90 days | Security and operations |
| Copies of log records held for delivery recovery | 30 days | Operations |
| Traces | 30 days | Operations |
| Network and security logs | Firewall logs: 90 days. Load balancer and www.parliview.org access logs, where enabled: up to 90 days | Security |
| Backups of our databases | 90 days. Point-in-time recovery copies: 35 days | Recovery after a failure |
| Session data | 5 days from your last activity | Operation of the Platform |
| Summaries, and records of corrections and withdrawals | A Summary is kept while it is shown; earlier versions are not kept. When we correct or withdraw a Summary we record the date and the reason and keep that record for 3 years | Accountability for published content, including the limitation period for defamation claims |
| Survey responses | Study end plus 5 years | Approved by the ethics committee |
| Future research contact email | Until you withdraw it | Consent-based |
| Messages to us, data protection requests and the record of withdrawals and objections | 3 years after the matter is closed | Evidence of how requests were handled |
| Anonymised datasets in research repositories | Indefinitely | Open science; a condition of the grant. Anonymised data is not personal data |
| Parliamentary records about MEPs and others (section 14) | While the Platform operates, reviewed annually | Public-domain source data |
The Principal Investigator reviews this schedule each year with UCD’s Data Protection Officer. Anything kept after study end must be supported by a documented assessment of why it is still needed.
What remains after withdrawal or deletion. When you withdraw, we delete the live records through our deletion procedure. When you remove a conversation from view yourself, it is hidden at once; the stored record is deleted when you withdraw or at study end. Copies in backups expire within 90 days. Entries in application logs expire within 90 days and are not deleted individually. Moderation records and acceptance records are kept for their own periods. Pseudonymised research datasets already extracted are kept as explained in section 11.
We use technical and organisational measures appropriate to the risks, including:
No system is completely secure. Please use a strong, unique password and turn on multi-factor authentication.
If a personal data breach occurs, we will assess it promptly. Where required, we will notify the Data Protection Commission within 72 hours of becoming aware of it. We will also tell you without undue delay if the breach is likely to result in a high risk to your rights and freedoms.
Taking part is voluntary. You can stop at any time, without giving a reason.
Stopping use. You can stop using ParliView at any time. To withdraw from the Study and have your account closed, email privacy@parliview.org. If you only stop using ParliView, your account and data stay until the end of the Study on 31 December 2028. They may be included in research datasets in the meantime.
Closing your account. To have your account and live data deleted, and to end research use of your data, email privacy@parliview.org and ask us to close your account. We then delete your account, conversations, feedback, request history and usage records from the live Platform, following our written procedure. We make no future research use of your data.
Objecting to research use only. You can object to the analysis of your data for the Study (purpose 2 in section 4) and keep using ParliView. Tell us at privacy@parliview.org. We then stop including your records in research datasets, and your account stays open. If you object to the processing we need to run the Platform (purpose 1), we cannot keep your account open and will close it as described above.
Records already in research datasets. We extract research datasets from time to time during the Study. If you withdraw or object before an extraction, none of your data enters a research dataset. If your records were extracted before you withdrew, we assess whether removing them would make the Study’s results impossible to verify or would seriously impair the Study. Where it would not, for example where the dataset has not yet been used in an analysis, we remove them. Where it would, the law allows us to keep them: Article 17(3)(d) GDPR for erasure, and Article 21(6) GDPR for objection. We tell you why. Kept records carry a research code, not your name, email address or user ID, and are used for no new analysis. We tell you this before you agree to take part.
| Right | What it means for you |
|---|---|
| Access | Ask whether we hold personal data about you and receive a copy, with the information in this notice |
| Rectification | Ask us to correct inaccurate personal data, or complete incomplete data |
| Erasure | Ask us to delete personal data, whether or not you withdraw. We delete it unless the law lets us keep it, and we tell you which ground applies. The grounds are: research datasets (section 11.1, Article 17(3)(d) GDPR); records needed to establish, exercise or defend a legal claim; and acceptance records, which are evidence of the information you were given and no longer link to you once your account is deleted |
| Restriction | Ask us to stop using your data, but keep it, while a dispute about it is resolved |
| Objection | Object on grounds relating to your particular situation. We assess each objection on its own facts; section 4 explains what happens to your account |
| Portability | Receive, in a machine-readable format, the data you gave us under consent: survey answers and your future-contact email. This right does not extend to data we process under public task, but you can still ask for a copy through access |
| Withdraw consent | Withdraw consent for surveys or future contact at any time. This does not affect processing already carried out |
| Complain | Complain to UCD, the Data Protection Commission or the authority where you live (section 18) |
How research limits apply. Irish law (section 61 of the 2018 Act) lets a research organisation refuse a request for access, rectification, restriction or objection in limited cases. It applies only where meeting the request would make the research impossible or seriously impair it, and only to the research part of the processing. It does not cover erasure; the erasure limit is explained in section 11.1. In plain terms:
Email privacy@parliview.org from the address on your account, or write to the postal address in section 1. Tell us which right you are using. There is no fee.
We may ask you to confirm your identity, usually by replying from the email address on your account or by signing in. We acknowledge requests within 5 working days and answer within one month. For complex requests, or if we receive many at once, we may take up to two further months; we will tell you within the first month if so. Requests are handled by a documented manual procedure; we do not have self-service deletion or export tools.
You can also contact UCD’s Data Protection Officer at gdpr@ucd.ie at any time. You can complain to the Data Protection Commission, or to the supervisory authority in the EU or EEA country where you live or work (section 18).
We do not make decisions about you based solely on automated processing that have legal effects on you or affect you in a similarly significant way.
www.parliview.org sets no cookies, stores nothing on your device and uses no analytics. Fonts are hosted on our own servers.
app.parliview.org uses only the cookies and browser storage needed to provide the Platform you have asked for. Irish law (Regulation 5(5) of S.I. No. 336 of 2011) does not require consent for storage that is strictly necessary to provide a service you have requested, so we do not show a cookie banner.
| Name | Type | Purpose | Set by | How long it lasts |
|---|---|---|---|---|
sid | Cookie (HttpOnly; Secure; SameSite=None) | Identifies your session. Your sign-in tokens and basic profile are held on our server, not in the cookie | ParliView | 5 days from your last activity; each use extends it. Set on your first visit, before you sign in |
| Amazon Cognito sign-in cookies | Cookies on our sign-in page domain | Sign-in, protection against forged requests, language choice and moving between sign-in pages | AWS (Amazon Cognito), for us | For the length of your sign-in session |
| Google sign-in cookies | Cookies on Google’s domains | Signing in to your Google account, if you choose Google sign-in | Set by Google; see Google’s cookie information | |
chakra-ui-color-mode | Browser local storage | Remembers whether you chose the light or dark theme | ParliView | Until you clear it |
pendingEventCode, pendingGroupUuid, pendingGroupId | Browser session storage | Carry an event code or invitation through sign-in so that you join the right group | ParliView | Deleted after you join the group or when you close the tab |
parliview_bookmarks, parliview_a11y_settings | Browser local storage | Your bookmarked messages and accessibility settings, kept on your device only | ParliView | Until you remove them or clear your browser storage |
We do not use advertising cookies, third-party analytics, tracking pixels or fingerprinting.
Source documents you open. The app sends a report to our servers when you open a source document from the Platform (section 3.2). The report is not needed to show you the document. We collect it for the Study, because whether participants check sources is part of what the Study measures. You are told about these reports before you agree to take part. To object, email privacy@parliview.org. We then exclude your records from research extracts, and your access is not affected. The app still sends the reports until you withdraw (section 11).
MEP photos. MEP photos in ParliView are loaded by your browser directly from the European Parliament’s servers. When this happens, the European Parliament receives your IP address, browser details and the address of our site. The European Parliament handles this data under its own data protection notice.
This section is for Members of the European Parliament (MEPs), former MEPs, and other people named in EP Material or Parltrack Data. That includes speakers in plenary debates, authors of parliamentary questions, and the Commissioners and other office-holders who answer them. We did not obtain this information from you. We cannot contact every person named in parliamentary records individually: there are several thousand, and the effort would be disproportionate. Publishing this section is the appropriate measure under Article 14(5)(b) GDPR.
| Category | Details |
|---|---|
| Identity and profile | Full name; date and place of birth; gender; the address of the official photo on the European Parliament’s website; whether the person is a current MEP |
| Roles and affiliations | Political group, national party, country, committee and delegation memberships, and other parliamentary roles, with dates |
| Parliamentary activity | Speeches, with the speaker’s name; written questions and answers; reports and other documents authored; roles on legislative files; individual roll-call votes |
| AI Output | Summaries of debates and other parliamentary business that name people and describe what they said, and Answers that participants receive about them. Every Summary is labelled AI-generated and is not reviewed by a person before it is shown |
| Corrections | Reports we receive about AI Output concerning a person, and what we did about them |
Some of this information reveals political opinions, such as party membership and votes.
We use this information to give participants sourced information about the work of the European Parliament: answering questions, preparing Summaries and vote breakdowns, and linking to the official record. Conversations about MEPs may also be analysed in pseudonymised form as part of the Study.
Text under the heading “Key points (AI paraphrase, not a quotation)” is the AI system’s description, in its own words, of what a speaker said. It is not a quotation, even where it resembles the record. We do not present AI paraphrase as quotation. If a Summary contains quotation marks around text that does not match the European Parliament’s record, please report it to corrections@parliview.org and we will correct it. To see what a speaker actually said, follow the link to the verbatim report. We ask participants not to republish a Summary as if it were a quotation.
Participants see profile cards, vote breakdowns, Summaries and Answers. AWS processes the data for us, including the AI models in section 6. We analyse it at UCD within research datasets. Our project partners receive only anonymised and aggregated results (section 7). The information is held in Ireland, and AI processing takes place in the EU. Profile, role and activity data is kept while ParliView operates, because it forms part of the public record of Parliament’s work; it is updated from the sources and reviewed annually. A Summary is kept while it is shown; earlier versions are not kept. When we correct or withdraw a Summary, we record the date and the reason and keep that record for 3 years.
ParliView is for adults. You must be 18 or over to ask for access, and the Participant Terms and Information say so. We do not knowingly collect personal data from anyone under 18. If we learn that an account belongs to a person under 18, we will close it and delete the data as described in section 11.
We may update this notice when the Platform, the Study or the law changes. When we publish a new version, you will be asked to confirm that you have read it before you next use the Platform; the version and date appear at the top. For changes that affect your rights or how your data is used, we will explain what has changed and why, on the Platform or by email, before the new version takes effect. Earlier versions are available on request from privacy@parliview.org.
This section applies in addition to the rest of this notice if you are in the United Kingdom.
Legal bases under UK law. UK law does not let a foreign university rely on “public task”. For you, our legal basis for both purposes in section 4 is therefore legitimate interests (Article 6(1)(f) UK GDPR). Our interest is conducting publicly funded research on how people use parliamentary information, in the public interest. We consider that this does not override your interests or rights because:
Consent (Article 6(1)(a)) applies to surveys and future contact, and legal obligation (Article 6(1)(c)) to requests we must meet by law.
Political opinions. For questions that may reveal political opinions, we rely on Article 9(2)(j) UK GDPR with paragraph 4 of Schedule 1 to the Data Protection Act 2018 (research in the public interest). That processing is carried out in accordance with the safeguards in Article 84B UK GDPR. Our safeguards are those in sections 4 and 5.
Your rights. The rights in section 11 apply, with the UK’s research limit. We may refuse a request for access, rectification, restriction or objection only where meeting it would prevent or seriously impair the research. That limit comes from Schedule 2, Part 6, paragraph 27 of the 2018 Act, and it applies only to research datasets. We make no significant decisions about you based solely on automated processing (Articles 22A to 22D UK GDPR).
Document views. UK law on information sent from your device requires clear information and a simple, free way to object to reporting for statistical or research purposes. Section 13 gives the information. To object, email privacy@parliview.org. We then exclude your records from research extracts, and your access is not affected.
Complaints. You can complain to UCD by emailing privacy@parliview.org. We will acknowledge your complaint within 30 days and tell you the outcome. You can also complain to the Information Commission, the UK regulator, at any time.
Transfers. Your data is transferred from the UK to Ireland, in the European Economic Area. UK law treats transfers to the EEA as approved (Schedule 21 to the 2018 Act and Article 45A UK GDPR). The transfers in section 8 to AWS global services take place under Standard Contractual Clauses; if you sign in with Google, Google acts for itself.
First, tell us. Email privacy@parliview.org (data protection) or support@parliview.org (the Platform). We will acknowledge your complaint within 5 working days and reply within one month.
UCD’s Data Protection Officer. You can raise a concern directly with UCD’s Data Protection Officer at gdpr@ucd.ie.
The Data Protection Commission. You can complain at any time to Ireland’s supervisory authority:
Data Protection Commission, 6 Pembroke Row, Dublin 2, D02 X963, Ireland
www.dataprotection.ie (complaints are made through the online form)
Your home authority. If you live or work in another EU or EEA country, you can complain to the supervisory authority there instead, and it will work with the Data Protection Commission. If you are in the UK, see section 17.