AI ethics in international collaborations

[TheresaWillem]

Title

AI ethics in international collaborations

Description

AI ethics is a difficult topic, even more so when we work in international and perhaps intercultural and/or interdisciplinary settings. In this session, we want to talk about the AI ethics challenges we face in international collaborations. We want to hear your experiences and workshop potential guidelines and principles that can be an orientation when back at home at our desks. Driving questions might be:

• Which values are important when working with AI (in general)?
• Which challenges have you faced in international AI projects?
• What baseline principles do we want to advocate for in international collaborations? How can we do that?

We will start our session by facilitating an open brainstorming on AI ethics in international collaborations. If sub-themes emerge we will split into smaller groups, world-café style.

We look forward to learning from your experiences!

Organizational

Host(s)

Christoph Feest (christoph.feest@helmholtz-munich.de), Theresa Willem (theresa.willem@helmholtz-munich.de)

Format

Brainstorming (and world-café, if feasible)

Timeframe

ca. 1 hour

Number of participants

1-40

Material

Beamer, flipchart, markers, post-its

[helenehoffmann]

@TheresaWillem

Thank you so much for hosting a session at this Unconference! Your participation is what makes this event truly special. Please, be prepared to give a 1-minute pitch for your topic to share at the beginning of the Unconference.

We’re excited to see you there and can’t wait for an amazing event! Best, The Organizing Team

[SusanneWenzel]

Dear @TheresaWillem ,

many thanks for your contribution! We hope your session was successful and constructive!

To make this not only a nice experience during the conference but a sustainable format, we would like to release a short result on the Helmholtz AI Website. Please provide us with a little report about the discussion in your session:

• One or two intro sentences about the topic and the respective format of your session
• What was most controversial, or about what did all participants agree on?
• What were the key ideas?
• Do you see any follow up on this discussion?

Depending on the intensity of the session it can be shorter (1/3 page) or longer (one pager).

Please post your report here by end of next week. Ina already confirmed to release that at a HAICON24 subpage at Helmholtz.ai.

Best regards, @helenehoffmann and Susanne

[TheresaWillem]

AI ethics is a difficult topic, even more so when working in international, perhaps intercultural, and/or interdisciplinary settings. This session discussed the ethical challenges we face in international AI collaborations. We started the session by reflecting on the challenges in our own collaborations. Then we shared them in the plenary, discussing which values the challenges touch upon and ideas that can be an orientation to alleviate the challenges when back at home at our desks.

AI ethics brings challenges as a technology, but often their alleviation is not within our scope of action. Therefore, our session included a discussion of research ethics and ethical challenges in international collaborations per se, which we believe are a breeding ground for techno-ethical challenges that surround the methods we publish and the products we build.

Key values we identified:

[incomplete list, in no particular order]

• Responsibility
• Risk reduction
• Privacy
• Transparency
• Fairness, including fair distribution of risks and benefits of (AI) technologies
• Inclusion / Diversity
• Respect, including respect for authorship criteria
• Environmental consciousness

Some guidance we found useful

For technical work:

• Think from the patient outcome – how will patients benefits? And who exactly are the patients we think of? If you can: write down which values you would like to embed in your model and how you can operationalize them (and keep checking whether you still agree with your values as the project progresses).
• Ask where the data comes from – and who collected them from whom. Be respectful and include data collectors in your papers (co-authorship or acknowledgement – whatever is feasible)
• Report on environmental impact of your models, use established frameworks (e.g., model cards for model reporting)
• When collecting data, install appropriate informed consent procedures that capture the cases the data might be used for. Your IRB / ethics committee can help with that.
• Properly license your contributions.

For people management:

• Install a data management and authorship plans from the beginning of your project, to keep things fair.
• When onboarding new members, include onboarding them regarding values and norms in your team/project. It will make everyone’s life easier when these things are talked about.
• Be conscious of other’s vulnerabilities.
• Make room to deal with intercultural issues in your projects.
• If you can, organize soft-skill trainings to help your group synchronize on values and grow together as a team that treats each other nicely and works towards the same ethical goals.

As a follow up and general advice we discussed as a group: Listen to your inner moral compass. If something seems off, ask for ethical assistance. You can always reach out to your IRB/ethics committee, your ombudsperson for good scientific practice, or our in-house ethicist. For example you can email theresa.willem@helmholtz-munich.de; Theresa works as an AI ethics consultant at Helmholtz Munich and will be happy to hear about what troubles you and see how you can navigate the situation. If deeper ethical analysis is required, you can book Theresa’s time, for example by applying for a (2-weeks or 6 months) consultants voucher – this will grant enough time to have a deep dive on your issue.