Purpose and scope
AI Safety Barcelona (AISBAR) is a community for people who want to understand and reduce the risks of advanced AI. These norms describe how we treat each other, so that newcomers and professionals alike can learn, collaborate and disagree safely.
They apply to everything run under the AISBAR name: our WhatsApp community, events and workshops, the website and member directory, and any other space where members meet through AISBAR.
Our values
- Respect. We treat every member as a capable adult, whatever their background, role or experience.
- People first. Before being right, it’s the people that matter. Listen, notice how someone is doing, and show empathy.
- Good faith. We assume the best intentions and ask before assuming the worst.
- Curiosity. Nobody is expected to know everything; we’re all here to learn.
- Inclusion. People of every discipline, nationality, gender, age and career stage are welcome.
What we expect
- Introduce yourself when you join, and welcome newcomers.
- Keep the WhatsApp group on topic: AI safety discussions, events, opportunities and community. Use the word "off-topic" when you step outside that.
- At events, leave space for others to speak and avoid dominating discussions.
- Answer beginner questions kindly.
- Respect people's attention. For instance, avoid tagging the whole group.
- A willingness and curiosity to learn more about AI safety and take committed action.
What is not acceptable
The following can lead to removal from AISBAR spaces:
- Harassment, bullying, intimidation or stalking, in person or online.
- Discrimination or insults based on gender, sexual orientation, ethnicity, nationality, religion, disability, age or any other personal characteristic.
- Unwelcome sexual comments or attention, including private messages after someone has said no.
- Sharing someone's private information without their permission.
- Spam, unrelated commercial promotion, and mass unsolicited messages to members, including messages built from the member directory.
- Threats, or advocating violence or illegal action against any person or organisation, including AI developers and researchers. AISBAR does not endorse either.
- Impersonating another person, or deliberately spreading false information.
How we discuss AI safety
AI safety questions are uncertain and often heavy. To ensure discussions are kept honest and kind, we recommend the following habits.
- Disagree openly. Say how confident you are. Saying that you don't know something is perfectly valid.
- Steelman first. Restate the strongest version of a view before you criticise it.
- No pressure. Members hold a wide range of views on AI risk. Do not push anyone to adopt yours or to change career.
- Look after each other. If a conversation is distressing someone, slow down and be kind, rather than escalating.
- Mind information hazards. Do not post detailed instructions that could enable serious misuse, for example in biology or cybersecurity. If unsure, ask an organiser first.
- Separate claim from opinion, and link sources when you can.
- Speak for yourself. Only organisers speak for AISBAR. Members' posts are their own views.
Privacy among members
- Do not share other members' phone numbers, emails or details outside the community without their permission.
- Do not post screenshots of the WhatsApp group elsewhere.
- Ask before photographing or recording people at events, and respect anyone who opts out.
- If you find a member through the public directory, contact them individually and for a relevant reason. Do not use it for mass outreach.
- Information in the member directory is there so members and AI safety people can connect. Do not copy it, scrape it or add people to lists or tools without their permission.
How AISBAR itself handles your data is described in the Privacy Notice.
Reporting a concern
If you experience or see behaviour that breaks these norms, write to aisafetybcn [at] gmail.com, to David Bravo (Director) at d.bravocomas [at] gmail.com, or to Elena Tairova (Partnerships & Ecosystem Lead) at tairova.e [at] gmail.com.
When we receive a report we will:
- Acknowledge it within 4 days.
- Listen to the people involved and keep the report confidential, sharing it only with those who need to know.
- Decide a response proportionate to what happened: a private conversation, a warning, a temporary mute or suspension, or removal from the WhatsApp community, from events and from the public directory.
- Tell the person who reported what was done, as far as privacy allows.
Anyone affected by a decision can ask for it to be reviewed by an organiser who was not involved. If someone is in immediate danger, call 112.
Changes and acceptance
By ticking the community norms box on our forms, you confirm that you have read these norms and agree to follow them. Organisers may update them as the community grows, and will announce significant changes by email or in the WhatsApp community. The latest version is always on our website.