Summary contributed by Victoria Heath (@victoria_heath7), Communications Manager at Creative Commons
*Authors of full paper & link at the bottom
Mini-summary: What if tech companies dedicated as much energy and resources to hiring a Chief Social Work Officer as they did technical AI talent? If that was the case, argues Desmond Upton Patton (associate professor of social work, sociology, and data science at Columbia University, and director of SAFElab), they would more often ask: Who should be in the room when considering âwhy or if AI should be created or integrated into society?â By integrating âsocial work thinkingâ into the process of developing AI systems, these companies would be better equipped to anticipate how technological solutions would impact various communities.
The code of ethics that guides social workers, argues Patton, should be used to guide the development of AI systemsâleading companies to create systems that actually help people in need, address social problems, and are informed by conversations with the communities most impacted by the system. In particular, before looking for a technical solution to a problem, the problem must be fully understood first, especially as itâs âdefined by the community.â These communities should be given the power to influence, change, or veto a solution. To integrate this social work thinking into UX and AI design, we must value individuals âbeyond academic and domain experts.â Essentially, we must center humanity and acknowledge that in the process of doing so, we may end up devaluing the power and role of the technology itself.
Full summary:
What if tech companies dedicated as much energy and resources to hiring a Chief Social Work Officer as they did technical AI talent (e.g. engineers, computer scientists, etc.)? If that was the case, argues Desmond Upton Patton (associate professor of social work, sociology, and data science at Columbia University, and director of SAFElab), they would more often ask: Who should be in the room when considering âwhy or if AI should be created or integrated into society?âÂ
By integrating âsocial work thinkingâ into their process of developing AI systems and ethos, these companies would be better equipped to anticipate how technological solutions would impact various communities. To genuinely and effectively pursue âAI for good,â there are significant questions that need to be asked and contradictions that need to be examined, which social workers are generally trained to do. For example, Google recently hired individuals experiencing homelessness on a temporary basis to help collect facial scans to diversity Googleâs dataset for developing facial recognition systems. Although on the surface this was touted as an act of âAI for good,â the company didnât leverage their AI systems to actually help end homelessness. Instead, these efforts were for the sole purpose of creating AI systems for âcapitalist gain.â Itâs likely this contradiction would have been noticed and addressed if social work thinking was integrated from the very beginning.
Itâs especially difficult to effectively pursue âAI for goodâ when the field itself (and tech more broadly) remains largely racially homogenous, male, and socioeconomically privileged; as well as restricted to those with âtechnicalâ expertise while other expertise is largely devalued. Patton asks, âHow might AI impact society in more positive ways if these communities [e.g., social workers, counselors, nurses, outreach workers, etc.] were consulted often, paid, and recognized as integral to the development and integration of these technologiesâŚ?â
Patton argues that systems and tools can be used to both help a community, and hurt it. âI havenât identified an ethical AI framework,â he wrote, âthat wrestles with the complexities and realities of safety and security within an inherently unequal society.â Thus, an AI technology shouldnât be deployed in a community unless a âmore reflective frameworkâ can be created that âprivileges community input.â When developing these systems, itâs important to admit, as Patton does, that the technical solution may not be whatâs needed to solve the problem.
Through his work at SAFElab, Patton has nurtured collaboration between natural language processing (NLP) and social work researchers to âstudy the role of social media in gun violence,â and create an AI system that predicts aggression and loss. Their approach was to first collect qualitative data by social workers trained in annotation who provided an analysis that then influenced the development of the âcomputational approach for analyzing social media content and automatically identifying relevant posts.â By working closely together, the social workers and the computer scientists were able to develop a more contextualized technical solution to a problem that was cognizant of the âreal-world consequences of AI.â
In order to effectively ask the right questions and deal with the inherent complexities, problems, and contradictions with developing âAI for good,â we need to change who we view as âdomain experts.â For the project at SAFElab, for example, they developed an âethical annotation processâ and hired youth from the communities they were researching in order to center âcontext and community voices in the preprocessing of training data.â They called this approach Contextual Analysis of Social Media (CASM). This approach includes collecting a baseline interpretation of a social media post from an annotator who provides a contextualized assessment; and then debriefing, evaluating, and reconciling disagreements on the labeled post with the community expert and the social work researcher. Once done, the labeled dataset is then given to the data science team to use in training the system. This approach eliminates the âcultural vacuumâ that can exist in training datasets from the beginning and throughout the entire development process.
The code of ethics that guides social workers, argues Patton, should be used to guide the development of AI systemsâleading companies to create systems that actually help people in need, address social problems, and are informed by conversations with the communities most impacted by the system. In particular, before looking for a technical solution to a problem, the problem must be fully understood first, especially as itâs âdefined by the community.â These communities should be given the power to influence, change, or veto a solution. To integrate this social work thinking into UX and AI design, we must value individuals âbeyond academic and domain experts.â Essentially, we must center humanity and acknowledge that in the process of doing so, we may end up devaluing the power and role of the technology itself.
Original paper by Desmond Upton Patton: https://dl.acm.org/doi/10.1145/3380535