Policy
Artificial intelligence is changing how organisations like ours design programs, communicate with participants, and process information. We use AI ourselves — carefully, and in specific ways — and we are taught and challenged by our participants, who are themselves grappling with what AI means for their cities.
Because AI carries real risks alongside its real benefits, we believe an organisation working with city leaders has a duty to be open about how it uses these technologies, where it limits them, and what it asks of others in return. This policy is our attempt to be clear.
It is written for our participants, our partners, our accreditation bodies, our funders, and anyone else who wants to understand the principles by which we operate.
These are the principles that guide how we use, govern, and limit AI across the Global Mayors Academy.
No decision that materially affects a participant — admission, assessment, certification, or progression — is made by an AI system alone. AI may help us prepare, draft, summarise, or analyse, but the people accountable for our program are the people who decide.
We use AI to extend human attention, not to replace it.
When AI is part of how we deliver something — a recommendation, a summary, a piece of analysis, a chatbot — we say so. We do not present AI-generated material as the work of a person, and we do not hide behind technical language to obscure how a system works.
If a participant ever wants to understand how AI was involved in something they have been shown or asked, we will tell them.
We do not feed participants' personal data, project work, or confidential cohort discussions into public, third-party AI systems. Where we use AI tools that process participant information, we use enterprise arrangements with appropriate data-protection terms, and we keep an internal record of which tools handle which categories of data.
Sensitive material from participants' cities — financial details, security-sensitive plans, identifying information about citizens — is treated with extra care and is not used to train any AI system.
AI systems can carry biases that affect leaders from different regions, languages, and backgrounds unevenly. Where we use AI in materials, recommendations, or evaluation, we test for these effects and we welcome feedback from participants who notice them. Our program is global; our use of AI must work for participants across the global North and the global South, not only for those whose context the underlying systems were designed around.
AI helps us produce more learning materials, support participants in more languages, and respond faster to questions across time zones. We use it to serve more leaders, more carefully — not to dilute the human relationships that are at the heart of what we do. Faculty contact, peer cohort work, and personal mentorship remain delivered by people.
The five commitments above are how we want to behave. The sections below describe what that looks like in practice today. We will update this section as our use of AI evolves.
We may use AI to help draft course materials, prepare summaries of group discussions, generate translations, suggest reading or peer-matching options, and respond to common questions through a program-support chatbot. Where we do, the materials are reviewed by a faculty member or a member of the GMA team before they reach participants.
We do not use AI to assess participants' work, to decide on admissions, or to issue certificates. Those decisions are made by people, with human accountability.
Participants' own project work — including their WOW★YOUR★CITY! projects — is treated as confidential intellectual property of the participant. We do not feed it into public AI tools, and we do not use it to train any AI model without the participant's explicit, written consent.
We may use AI to help draft initial copy, generate first-pass translations, and produce supporting visuals. All public-facing material is reviewed and edited by a human author before publication. We do not generate fabricated quotes from participants, faculty, or partner institutions, and we do not present synthetic imagery of real people.
If we use AI-generated imagery in a context where it might be mistaken for a photograph, we say so plainly.
We use AI tools to help with internal tasks such as scheduling, drafting documents, processing voice notes, and analysing structured data. Where these tools touch participant or partner information, we use enterprise-grade arrangements with appropriate data-protection terms. We maintain an internal register of which AI tools we use for which purposes — available on request to accreditation reviewers and major funders.
We do not paste confidential participant or partner information into consumer-grade AI tools.
Responsibility for this policy sits with the GMA leadership team, with input from our COO and our technology lead. We review the policy at least annually, and sooner if our use of AI changes meaningfully. Participants, partners, and members of the public can raise concerns or questions about our use of AI by writing to contact@globalmayors.academy, and we will respond.
Where a concern reveals a gap or a mistake in how we have applied this policy, we treat that as feedback that improves the next version. We expect to learn in public.
We ask that participants and partners apply the same care to AI use that we ask of ourselves. Specifically, we ask that:
These expectations are part of our broader Code of Conduct and are reinforced in the Terms and Conditions that govern program participation.
This policy is a living document. We review it at least once a year and update it whenever our use of AI changes in a way that affects participants, partners, or the public. Substantive changes are dated and noted in the version history at the bottom of this page.
We will not weaken the protections in this policy without making the change publicly visible.
If you have a question, a concern, or a suggestion about how we use AI, please write to us at contact@globalmayors.academy. A member of our team will respond.
Global Mayors Academy is a program of Living Cities Earth (Swiss Verein). Our Ethical AI Policy is published alongside our Terms and Conditions and Privacy Notice.