Session: Applying Artificial Intelligence in Collaborative Research to Solve Grand Challenges for Social Work (Society for Social Work and Research 29th Annual Conference)

Background and Purpose: Artificial Intelligence (AI) and its applications have become ubiquitous across multiple fields. AI has great potential to enhance social work practice through more efficient and tailored services. Collaborative research, especially that involving AI, is likely to have a strong impact on solving complex social problems. Such complex social problems include those articulated by the Grand Challenges for Social Work (e.g., racism, social isolation, unequal opportunity). Too intractable for social workers to solve alone, such problems call on social workers to collaborate in multidisciplinary teams using cutting-edge and powerful technology such as AI, while employing it ethically and responsibly. As such, the current symposium theme focused on highlighting the successful multidisciplinary collaborations that ethically and responsibly employ AI in solving complex social problems related to the Grand Challenges (i.e., racial bias within medical settings, social isolation and trauma, lack of adequate family services).

Methods: This symposium includes three presentations complementary in their data sources and methodologies. The first study employed an AI-powered mixed methods approach to examine 259,350 Reddit posts from 30,372 users across 10 years to examine the types of social support Reddit users with child abuse histories seek across multiple Reddit communities. The second study employed machine learning techniques to detect racial bias in clinical case notes from 2,467 nurses in Brazilian hospitals. The third study applied the PRISMA-ScR to a scoping review that conducted a comprehensive literature review, with approximately 7,000 articles examined across 10 years to understand the role of AI in collecting data, analyzing data, and rendering resources to couples, parents, and children in the context of family-centered services.

Results: The first study found that by connecting people with similar experiences, online communities help combat social isolation and support members who are healing from childhood trauma. The second study showed that AI-assistant technologies in hospital settings need to account for racial biases to ensure health equity and that collaboration with medical professionals is paramount in debiasing and providing human oversight of AI models. The third study demonstrated that identified studies commonly focused on maternal and adolescent populations, with various AI methods employed and many articles citing conceptual frameworks involving the ethics of AI.

Conclusion and Implications: This symposium fits well with the SSWR 2025 conference theme of Strengthening Social Impact through Collaborative Research because it showcases successful collaborations and co-creation of knowledge between social workers, data scientists, and medical professionals. Additionally, the symposium offers actionable recommendations to use AI ethically and responsibly in the context of serving individuals, families, and communities. Key contributions of the symposium include critical reflection on (1) harnessing AI for social good and its alignment with the Grand Challenges; (2) building successful collaborations across multidisciplinary fields; and (4) reducing bias in AI application to eliminate the perpetuation of racism, inequality, and oppression. The discussant, with expertise in applying data science and machine learning fairness to promote family well-being, will contribute a translational component that speaks to social workers leveraging collaborative AI research to solve the Grand Challenges.