AI Transparency Statement

1.Purpose of this statement

Because machine learning is built into several parts of the platform, we describe here — in one place and in plain terms — where it operates, whose systems perform it, what its limits are, and how people stay in control. We update this statement when the platform’s use of AI changes materially.

2.Where AI is used

• Submission assistance. When you paste an announcement or upload an abstract, a language model extracts the structured fields (title, dates, speakers, location) to pre-fill the form. You review and can correct everything before submitting.
• Summaries and descriptions. Some descriptive texts on the platform — content summaries, topic overviews, extracted keywords — are produced with AI assistance from the underlying scholarly material.
• Classification. Content is assigned to scientific domains and topics by a model so that browsing and filtering work across disciplines.
• Semantic search and recommendations. Text — content, search queries, and (if you maintain one) your profile — is converted into numerical representations (embeddings) used to match related work, rank search results, and populate recommendation surfaces.
• Moderation assistance. Submissions are screened by a model for policy violations before human review.
• Decorative imagery. Abstract banner and thumbnail images for scientific domains are AI-generated artwork; they depict no real people or events.

3.What AI is not used for

• No decisions producing legal or similarly significant effects are made solely by automated means (Article 22 GDPR) — moderation outcomes involve human review.
• No chatbot or conversational agent interacts with you on the platform.
• No emotion recognition, biometric categorisation, or identification of natural persons.
• No advertising profiles, and no sale of data to AI vendors.
• No first-party transcription of audio or video runs on the platform today; if that changes, this statement and the Privacy Notice will be updated first.

4.Provider and data handling

All language-model, embedding, and image generation runs through the API platform of OpenAI (contracting entity for the EEA: OpenAI Ireland Ltd), acting as our processor. Under OpenAI’s API terms, content submitted through the API is not used to train OpenAI’s models unless the customer opts in. Under standard API controls, abuse-monitoring logs and some Responses API application state may be retained for up to 30 days. Some calls explicitly disable application-state storage, but that setting is not yet uniform across every World Wide AI feature. OpenAI’s data processing addendum uses an adequacy decision or the EU Standard Contractual Clauses, as applicable, for onward transfers.

We send only the text a task needs: the content being processed, the search query being matched, or the profile text being indexed. We do not send credentials, analytics event records, or data unrelated to the task. A document or profile can itself contain an email address or other personal data, so review material before invoking a feature. Model usage is metered and bounded by spend controls on our side.

5.Machine-generated content and its limits

Language models produce plausible text, not verified fact. Machine contributions on World Wide can therefore be incomplete, outdated, or wrong — a summary can flatten a nuance, an extracted date can be misread, a classification can miss the field. Treat machine-assisted material as a pointer to the original source, and verify details that matter — deadlines, eligibility rules, scientific claims — against that source. Nothing machine-generated on this platform constitutes professional, legal, medical, or career advice.

We are extending clear “machine-generated” marking across all AI-produced elements of the interface. If you find machine-generated text that is wrong, misleading, or not identifiable as such, report it (Section 7) and we will correct or label it.

6.Human oversight

People remain in the loop at the points that matter: submitters review pre-filled fields before anything is published; moderation outcomes are human-reviewed; editorial corrections override machine output; and every AI integration runs behind monitoring and spend limits so that a misbehaving model degrades to the non-AI experience rather than taking the platform with it.

7.Questions and reports

Questions about AI on World Wide, reports of incorrect or unlabelled machine-generated content, and objections to AI-assisted processing of your data (see the Privacy Notice, Section 10) all reach us at info@world-wide.org.