Fostering the discourse about gender aspects in AI technologies

By Carolin Penke

For International Women’s Day 2024, we organized an event revolving around issues concerning women and other marginalized groups in the rapidly developing AI landscape. This included a talk and an open discussion. The interest was huge and you can read up on it in our blog post.

In fact, the interest was so enormous, that we were invited at multiple occasions to present this important topic and create spaces for an open exchange and discussions. This included  (a) a World Café at the Helmholtz AI conference 2024 HAICON24 and (b) a talk at the public-facing event series “Wissenschaft Online” of Forschungszentrum Jülich. A recording of this event can be found here.

The World Café “Gender and AI” at HAICON

The organizers of the Helmholtz AI conference realized that discussions and exchange were central aspects of a successful conference. To this end, they implemented the World Café format, where several round tables are set up. Each table hosts a discussion for a specific topic, facilitated by a moderator. For each of the three 20 minute sessions, participants could choose a topic of interest, switching tables in between, while notes in form of a large mind map were kept at each table.

The Gender and AI World Café session focused on the intersection of gender issues and artificial intelligence. The discussion aimed to address the societal dangers of recent AI advances, particularly for women and minorities. Bias in AI models was discussed, as well as diversity and inclusion in the AI workplace, and how the two are related.

All participants agreed that a diverse and inclusive workforce is key to attracting the best talent, gathering a variety of perspectives, fostering creativity, and preventing blind spots in the development of AI solutions, as well as in science in general. We should not tire of pointing out the benefits of diverse and inclusive work cultures, not only because they are more fair and equitable, but also because they lead to better team performance and better research.

Many different ideas and aspects were discussed on how to achieve a better gender balance in the field, e.g. through role models, outreach activities in schools and by raising awareness. Open questions were how to reward and value activities in this area within academia and how the German situation differs from other countries. The Helmholtz AI conference itself was identified as a positive example, with a high number of women attendees and speakers and inclusivity measures such as childcare.

One of the major problems with AI systems in production has been identified as unchecked bias in models related to sex, gender, or other attributes. Bias cannot be completely avoided and can be introduced at every step of the model lifecycle, from data collection to deployment. The Helmholtz AI conference highlighted some interesting contributions in the area of bias detection and mitigation. In addition to ongoing research in this area, it is important to educate the general public as well as researchers to be aware of the limitations of AI and to be sensitive to potential problems. Generative AI poses a unique risk due to its lack of transparency and our tendency to humanize its output. Clear rules and incentives for responsible AI development are needed and must be enforced. The European AI Act can be a step in the right direction. As a concrete outcome, we identified the need to train researchers in the general area of AI ethics, for example in the form of a courses.

Wissenschaft online: „Was soll schon schiefgehen? Gefahren von KI aus einer geschlechtergerechten Perspektive“

Wissenschaft online is a weekly public event established by Forschungszentrum Jülich, where selected researchers present their work to an online audience, followed by the opportunity to ask questions and engage in a discussion.

After the successful event at International Womens’ Day, I was asked to present the topic once more in German. In the fast-paced world of AI, of course, updates were needed to present the most current developments.

Conclusions

Across all these events, the high level of audience engagement underscored a keen interest in understanding the inner workings of AI, its transformative potential and its challenges. Participants were open to a nuanced exploration of its societal and ethical implications. Raising awareness for these was the main goal for this series of talks, and judging from the thoughtful discussions and the feedback we received, this was achieved. We examined multiple facets of the technology, and emphasized that it can also be a powerful and inclusive tool for progress, when developed responsibly.

Moving forward, we hope these conversations inspire continued dialogue and action to ensure AI serves everyone equitably and responsibly.

About the blog post author:

Carolin Penke is a researcher at Forschungszentrum Jülich, Germany, where she optimizes and facilitates the training of large GPT-like language models on GPU-accelerated clusters. Her doctoral studies in mathematics involved the development of efficient algorithms for structured eigenvalue problems in electronic structure theory. Currently, her research applies tools from numerical linear algebra to enhance the efficiency of AI training.