Privacy Notice
Data Annotation Platform
1. Who we are and what this notice covers
QCRI ("we", "us", or "our") operates micromappers.qcri.org (the "Platform") and determines how personal data covered by this notice is used. This notice explains how we collect, use, share, protect, and retain personal data when you register, perform annotation tasks, submit recordings, or communicate with us.
This notice accompanies our Terms and Conditions and any task-specific information or consent forms. It does not itself grant intellectual property rights or replace a recording consent form.
2. Data we collect
Account and contact details: Your name, email address, account identifier, authentication information. These are your personal data.
Contributions and task records: Labels, annotations, judgments, written responses, transcripts, feedback, submitted recordings, task identifiers, submission times, and quality-review records. These are not your personal data. These are the data that you are participating either paid or voluntarily.
Recordings and related information: For tasks you undertake with the relevant authorization, your voice and audio, and, where the task requires it, images or video. We may create transcripts, speaker labels, and task-relevant features from these recordings. Your voice or appearance may remain identifiable even after your name is removed.
Task profile information: Where relevant and explained in advance, information such as languages spoken, dialect, age range, and general country or regional background. We identify which fields are required and which are optional.
Technical and consent records: Where collected, IP addresses, browser or device information, access and other related logs, and records of notices shown and choices made, including the notice version and consent timestamp.
The specific information collected for each project will be identified in its task notice. Do not include unnecessary information about yourself or other people in free-text answers or recordings.
3. How we obtain your data
We obtain data directly from you when you register, complete tasks, and uploaded recordings. We also collect the technical information described above through use of the Platform. Where an authorized annotation service provider supplies your registration details or task records, we will identify that provider and the information supplied.
Certain details, such as your name and email address, may be required to create and administer your account. Where information is necessary for a particular task, we explain this before collection. You may decline to participate in such tasks.
4. Why we use your data and our permitted reasons
Platform administration: We use your account details to manage access, assign work, communicate with you, review quality, prevent misuse, and maintain security. Our permitted reason is processing necessary for these specified lawful operational purposes, where permitted under Qatar law; where consent is required, we obtain it before processing.
Research and commercial development: Annotations, recordings, transcripts, responses, and related task data may be used for both research and commercial purposes. These include creating and validating datasets; conducting and publishing research; training, fine-tuning, testing, and evaluating AI models; and developing, improving, licensing, and commercializing datasets, software, technologies, products, and services.
These activities may include speech recognition, language understanding, speech synthesis, and multimodal AI, as specified for the relevant project.
Compliance and rights requests: We may process relevant records to meet applicable legal duties, investigate incidents, and respond to privacy requests or complaints. Any continued retention for these purposes must have a valid legal justification.
Account names and email addresses are used for administration and communication, not included in public research datasets or model-training material merely because you have an account. Any different proposed use would require its own explanation and applicable authorization.
5. Who may receive your data
Authorized personals: Personnel who need relevant information for platform administration, project delivery, quality assurance, security, or research.
Service providers: Approved hosting, storage, IT-support, annotation-management, and other providers processing data on our behalf under appropriate instructions and safeguards. Where necessary, an annotation service provider may receive limited account, participation, and quality information for coordinating the work.
Research collaborators and commercial partners: Relevant contributions may be shared or licensed for the research and commercial uses disclosed for the project. The information supplied, recipients or recipient categories, intended uses, and access conditions will be explained before the relevant sharing. Partners acting as independent data controllers will be identified as such.
Public access and authorities: We may publish anonymized findings or datasets assessed as suitable for public release. Identifiable recordings require the separate authorization described above. We may also disclose relevant information when lawfully required by a competent authority.
We do not publish annotator account names or email addresses as part of a research dataset. Sharing or licensing a dataset is limited by the disclosed purposes, applicable permissions, and privacy requirements; it is not permission for unrestricted use of personal data.
6. Storage locations and international access
Platform data is stored within QCRI’s cloud compute infrastructure.
7. How long we keep your data
We retain personal data only for as long as necessary for its disclosed purpose or a justified legal requirement. Our retention schedule is:
| Record category | Period or clear retention criteria |
|---|---|
| Account and profile data | May be closed after 24 months with prior notice. |
| Submission files and evaluation records | Up to 2 years after evaluation. Final submissions and essential evaluation records may be retained for up to 5 years. |
| Identifiable public leaderboard records | Up to 5 years after publication, unless earlier removal is required. |
| Terms acceptance and consent records | While applicable and up to 5 years afterward, including after consent withdrawal. |
At the end of the applicable period, we delete the information or anonymize it so that you can no longer reasonably be identified. Where continued retention is legally required, we restrict use to the purpose justifying retention. Replacing a name with an annotator or speaker ID does not, by itself, make data anonymous.
8. Safeguards
We implement appropriate security measures, including role-based access controls, restricted administrative access, encrypted connections (HTTPS), and secure server infrastructure. We investigate security incidents and notify affected parties where legally required. Security concerns can be reported to the platform management team.
9. Your rights and withdrawing consent
Subject to applicable law and any applicable exceptions, you may request information about processing, access to your personal data, correction of inaccurate information, deletion where the conditions for deletion are met, or object to certain processing. Where we rely on your consent, you may withdraw it.
Send requests or complaints to the platform management team We may request information reasonably needed to verify your identity and locate the records. We will assess and respond in accordance with applicable requirements and explain any limitations on the action we can take where required.
10. Children and personal data of a special nature
This general annotator notice is intended for adults aged 18 or over. A project involving children or personal data of a special nature requires separate project arrangements, appropriate notices and authorizations, and any permission required from the competent authority before processing. Such data may include information about health, ethnic origin, religious beliefs, or other categories protected by Qatar law.
Do not submit such information unless it is expressly requested through an appropriately authorized project. Please contact us if you believe this information has been collected unintentionally.
11. Cookies and automated quality checks
We use essential cookies for authentication, session management, security, and user preferences. Session cookies are retained for up to 14 days, while security and preference cookies may be retained for up to 12 months. Users can delete cookies through their browser settings.
12. Changes and contact
We will publish revisions with an updated version and effective date and communicate material changes through email. Before introducing a new purpose, we will provide the necessary information and establish the required basis for processing.