Pennings, A.J. (2026, Sep 02) Emerging AI Frameworks in Leading Science, Technology, and Society (STS) Schools and Centers apennings.com https://apennings.com/science-and-technology-studies/emerging-ai-frameworks-in-leading-science-technology-and-society-sts-schools-and-centers/

Introduction

Science, Technology, and Society (STS) frameworks remain critical for AI analysis and governance because they treat artificial intelligence as a sociotechnical system shaped by, and in turn reshaping, institutions, power, social relations, values, and material infrastructures, rather than as a purely technical object to be controlled through standards or rules alone.

Drawing on classic STS approaches and the emerging strategies visible in major STS programs, including those highlighted by me previously and active initiatives at MIT, Stanford, Harvard, Berkeley, Cornell, Stony Brook, and related centers), this overview maps the most relevant frameworks and shows how leading schools are operationalizing them for AI.[1]

Emerging AI Strategies in Leading STS Schools and Centers

The major US STS nodes such as MIT, Stanford, Berkeley, Harvard, Cornell, and Stony Brook (Department of Technology and Society in CEAS) are places where STS has long intersected with engineering, policy, and management. Recent developments show these programs actively shaping AI governance agendas:

MIT Program in Science, Technology, and Society continues its foundational role while engaging AI through ethics, journalism, and Institute-wide reflection on education and research norms. Faculty and events address governance, liability, and the societal embedding of AI, often in dialogue with the Schwarzman College of Computing.

Stanford Univeristy STS Program and HAI offers concentrations and courses that deconstruct AI hype, examine values built into systems, and link technical and social analysis. The broader Stanford ecosystem (HAI, Law AI Initiative, RegLab) operationalizes STS insights into human-centered AI, policy research, and governance tools, emphasizing interdisciplinary collaboration between STS, computer science, law, and management science.

Harvard Program on Science, Technology, and Society (Kennedy School) focuses on policy and institutional design. Recent work advances proactive, power-sharing approaches to AI governance that prioritize human flourishing, democratic stability, and economic empowerment over purely reactive risk management.

UC Berkeley’s CSTMS and AI Security Initiative at CLTC combines critical theory and social-justice orientations with practical risk-management standards development. Emphasizes multistakeholder processes, equity in benefit distribution, and contributions to national and international standards (including NIST-related efforts).

Cornell University’s Department of Science and Technology Studies + Global AI Initiative integrates global and pluralistic perspectives into AI research, design, evaluation, and governance. Focus areas include inclusive AI for diverse communities, transparency, accountability, and public oversight, with explicit attention to translating research into policy and practice.

Stony Brook University’s Department of Technology and Society has a new Department of Technology, AI and Society in the College of Engineering and Applied Sciences (CEAS). It is building on its long-standing TSM tradition and the motto that engineering is “much too important to be left to the engineers,” Stony Brook’s Department of Technology and Society is expanding into a dedicated Technology, AI and Society department with a new chair. NY State investment supports faculty hiring, new degree pathways, and research organized around ethics, equity, and justice, with applications to energy, health, and societal challenges. This represents a direct engineering-school pathway for STS-informed AI systems management and governance.

Other programs (Edinburgh, Twente, UC San Diego, etc.) similarly emphasize multidisciplinary training and critical engagement, reinforcing a global STS conversation on AI.

Implications for Governance

These frameworks and institutional strategies converge on several practical orientations. Governance must be process-oriented, participatory, and reflexive rather than purely compliance-based. Accountability should be traced across networks rather than assigned to a single locus. Technical knowledge claims and social ordering must be examined together. Context, power, and global diversity matter; one-size-fits-all technical standards are insufficient.

Engineering and policy education should integrate STS tools so that future designers and managers treat social acceptance, legitimacy, equity, and sustainability as core system requirements.

Leading STS programs are already institutionalizing these insights through new courses, concentrations, research initiatives, policy engagement, and (in cases such as Stony Brook) departmental reorganization around AI and society. The result is a maturing field in which classic STS analytics such as SCOT, ANT, co-production, sociotechnical systems, and RRI—are being refined and applied to the distinctive challenges of general-purpose, generative, and agentic AI.

Some Major STS Frameworks from Major Universities Applied to AI Governance

Social Construction of Technology (SCOT)

Different social groups interpret AI differently until “closure” stabilizes dominant meanings of safety, fairness, or intelligence. Governance processes themselves become sites of negotiation among developers, regulators, affected communities, labor, and Global South actors. Stanford’s STS courses (e.g., “Constructing and Deconstructing Artificial ‘Intelligence’”) explicitly train students to question hype and examine whose values are inscribed in systems.

Actor-Network Theory (ANT)

AI systems emerge from heterogeneous networks of human and non-human actants (algorithms, datasets, chips, standards, legal texts, users). Agency and accountability are distributed and can be displaced. Recent applications map how responsibility shifts in generative and agentic AI, medical AI, and public-sector deployments. This lens is especially useful for tracing why accountability often fails to land on the most powerful actors.

Co-production (Jasanoff and others)

Ways of knowing AI (benchmarks, risk assessments, capability evaluations) and ways of ordering society (who is protected, who decides, what counts as harm) are produced together. Governance frameworks do not merely regulate a pre-existing technology; they help constitute what legitimate AI is. Harvard’s Program on Science, Technology, and Society (housed at the Kennedy School) and related work on power-sharing liberalism exemplify this orientation, linking knowledge production to democratic and institutional design.

Sociotechnical Systems Thinking and Its Extensions

Classic joint optimization of social and technical subsystems has been updated for intelligent and agentic systems. Newer “intelligent sociotechnical systems” approaches examine how AI agents can themselves reconfigure coordination structures, creating recursive governance challenges. This resonates with engineering-school STS traditions that emphasize systems management.

Responsible Research and Innovation (RRI)

Anticipation, inclusion, reflexivity, and responsiveness provide a practical STS-informed governance orientation. Leading programs adapt these dimensions to AI through upstream engagement, continuous reflection, and adaptive institutions.

Complementary Lenses

Infrastructure studies reveal the invisible classification systems and data infrastructures that quietly govern outcomes. Feminist and standpoint STS highlight situated knowledges and care. Critical approaches question technological determinism and open possibilities for democratizing design and oversight. Environmental and justice-oriented STS (strong at Berkeley’s CSTMS) foreground energy, material, and equity impacts.

In short, STS does not merely critique AI governance; it supplies the conceptual and institutional resources for building more legitimate, adaptive, and equitable governance arrangements. The programs at MIT, Stanford, Harvard, Berkeley, Cornell, Stony Brook, and peer institutions demonstrate how these frameworks are moving from theory into curriculum, research agendas, and real-world policy influence.

Implications for AI Strategies, Research, and Technical Training

STS frameworks encourage AI strategies that are:

Research-informed and reflexive — Research agendas incorporate anticipation of social effects, diverse knowledge inputs, and ongoing evaluation of assumptions.

Technically rigorous yet context-aware — Technical training (programming, systems design, evaluation methods, machine learning fundamentals) is paired with skills in stakeholder analysis, ethical reasoning, implementation studies, and impact assessment.

Network and systems-oriented — Strategies address the full assemblage of data, models, infrastructure, organizations, and users rather than isolated technical components.
Inclusive and power-sensitive — Training and research design deliberately surface whose knowledge counts and who benefits.
Adaptive — Both research programs and educational curricula build capacity for continuous learning as AI capabilities evolve.

In practice, this produces AI strategies that integrate technical roadmaps with organizational change plans, research priorities with public value considerations, and skill development with critical literacy. Graduates and researchers trained in these environments are equipped to design, evaluate, and steer AI systems as sociotechnical endeavors.

Leading STS programs at MIT, Stanford, Harvard, Berkeley, Cornell, Stony Brook, and peer institutions demonstrate how classic frameworks such as SCOT, ANT, co-production, sociotechnical systems thinking, and RRI and how they are being translated into concrete research agendas, technical training pathways, and institutional AI strategies. The result is a maturing approach in which technical excellence and sociotechnical insight are pursued together as essential components of effective AI strategy.

Notes

[1] At the State University of New York, Korea, we offer the Stony Brook curriculum from the Department of Technology, AI, and Society in New York.
AI Prompt(s) Review AI strategies emerging in STS schools in MIT, Cornell, Stony Brook, Harvard, Stanford, Berkely and other schools mentioned in my previous research on the increasing importance of STS.

© ALL RIGHTS RESERVED

Not to be considered financial advice. AI is often used, and results are thoroughly interrogated. Links are used for some citations. Views are my own and do not express the stances of my employers, past or present.

---

Anthony J. Pennings, PhD is a Professor at the Department of Technology and Society, State University of New York, Korea and a Research Professor for Stony Brook University. He teaches AI and broadband policy. From 2002-2012 he taught digital economics and information systems management at New York University. He also taught in the Digital Media MBA at St. Edwards University in Austin, Texas, where he lives when not in Korea.

Related

Category: artificial intelligence, science and technology studies, sustainable development

Comments

Comments are closed.