10 to the 23 AI logo
Stephen Lieberman
Through 1023AI
AI in the Room — Codesigning AI Governance for Social Services

You belong in this room.

AI is reshaping the systems that serve your clients, your communities, and the people you have dedicated your career to. We are building the governance frameworks that make sure it does that well. Your knowledge, your experience, and your voice are what make that possible.

Apply to ParticipateLearn more about the initiative ↓

Artificial intelligence is already making consequential decisions about the people and communities human services exists to serve. Screening tools flag families for investigation. Algorithms recommend placements, prioritize cases, and allocate resources. Predictive systems shape who gets services and who does not. The stakes could not be higher, and the pace of deployment is accelerating.

You bring knowledge and experience that can only come from the field: what clients actually face, what communities actually need, and what accountability to vulnerable people looks like in practice. Getting AI governance right requires you at the table where these decisions are made.

AI in the Room is here to make sure that happens, and your participation is what makes it possible.

What We Are Building

AI in the Room is an initiative at the University of Southern California Suzanne Dworak-Peck School of Social Work, anchored within the Grand Challenge to Harness Technology for Social Good in partnership with NASW. It brings together the two communities that must work together to get AI governance right: the social work and human services professionals who understand what is actually at stake for clients and communities, and the AI researchers and technologists who understand how these systems are built from the inside.

The mechanism is a structured participatory codesign process built around what actually happens when AI meets real lives. This process creates AI governance policies that are iteratively designed, tested, and refined before moving through NASW into professional standards, practice guidelines, and national frameworks that reach the entire field.

When the people closest to this work help govern it, AI can become a genuine force multiplier for the communities we serve:

Support reaching families earlier, before a crisis hardens into a removal
Resources finding communities that systems have historically overlooked
Practitioners freed from administrative weight to do the human work only they can do
Deployed systems that answer to practitioners and communities rather than operating beyond their reach

Why Your Voice Is Irreplaceable

A governance framework is only as sound as the knowledge built into it, and no single group holds all of that knowledge. Each voice in the room sees what the others cannot.

Practitioners

You know what deployment actually does on the ground, where a tool helps, where it fails, and where it quietly reshapes the work. That knowledge lives in practice and nowhere else.

Community members and advocates

You know what it is like to be on the receiving end of these systems, to be sorted, flagged, or decided about by a tool you may never see. That is the reality others cannot see from where they sit, and governance that ignores it fails the people that need it most.

Leaders

You understand the policy, institutional, and organizational forces that decide whether a framework survives contact with real systems. Without that knowledge, governance is written for a world that does not exist.

AI researchers and technologists

You know how these systems actually work on the inside, what they can do, what they cannot, and where they break in ways no one outside the build ever sees. That knowledge is what keeps governance tied to the technological possibilities, and this work needs you in the room to provide it.

Whether AI becomes a tool for equity or another mechanism for harm will not be settled by technology alone. It depends on work like this, and on people like you choosing to do it.

Apply to Participate

About the Team

Stephen Lieberman

Founder, 1023AI · DSW Candidate, USC

Stephen Lieberman is a DSW candidate at the USC Suzanne Dworak-Peck School of Social Work, President of the NASW Southern California Caucus, and President of the Phi Alpha Honor Society, Omicron Epsilon Chapter. He brings more than 20 years of leadership in AI, technology, and social innovation across nonprofit, government, industry, and academic sectors, including senior roles at Northrop Grumman and the Naval Postgraduate School, where he served as a Principal Investigator for multiyear social technology research and development programs.

Dr. Marya Wright, DSW, LCSW

External Design Partner · President-Elect, NASW California

Dr. Marya Wright, DSW, LCSW is the President-Elect of NASW California, a USC DSW alumna, Phi Alpha leader, and a published researcher on algorithmic decision-making in child welfare. She is a nationally recognized social work educator, consultant, and forensic social worker. She serves as External Design Partner for this initiative, bringing deep domain expertise and the institutional knowledge to carry the work we do at AI in the Room design sessions forward into professional practice and national policy.

Apply to Participate

Join AI in the Room

AI in the Room is building a community of social workers, AI researchers, nonprofit leaders, policymakers, and community members to collaboratively shape responsible AI in social services. Through structured Design Labs and ongoing collaboration, participants will help inform the development of a Framework for Responsible AI in Social Services. Interest does not obligate participation, and all inquiries are confidential. We will respond personally to every inquiry.

Your brief answers to the two questions below help us bring together the right mix of experiences, insights, and voices to collaboratively shape responsible AI in social services. You can paste up to one page of text into these fields.

Areas of experience or expertise (optional — select all that apply)
Would you be interested in… (optional — select all that apply)

Fields marked * are required.