Call for Papers: ACM International Conference on Multimodal Interaction

Track: Blue Sky Track (A.K.A. Vision Track)

Date: November 12-15, 2026

Location: Shenyang, China

Website: http://icdm2026.neu.edu.cn/main.htm

Aims and Scope

The Blue Sky Track invites the research community to present visionary propositions that could inspire impactful directions and compelling new research opportunities for the ICDM community within the foreseeable future. Unlike traditional research papers, these submissions should:

• Expose gaps in current data mining research
• Propose bold, forward-looking ideas that tackle fundamental challenges
• Introduce unprecedented algorithmic or methodological opportunities
• Question prevailing assumptions in the community

The goal is to stimulate innovation and foster new research trajectories that will result in paradigm changes and enable a “quantum leap” in knowledge.

What is a Blue Sky Paper?

A Blue Sky paper presents a bold vision with high potential to expand the current ICDM research agenda by defining new research topics and transformative problem areas. Authors are encouraged to look beyond existing methodologies and technologies to present “out-of-the-box” thoughts.

What is NOT a Blue Sky Paper?

• Work proposing slightly better solutions to well-studied problems.
• Papers dominated by a specific solution with limited discussion on broader future impact.
• Applications of existing techniques to a specific domain without significant novel challenges for the ICDM community.

Why Submit?

Shape the Future: Influence long-term research visions and trajectories in the data mining community.

High Visibility: Accepted papers will appear in the conference workshop proceedings.

Recognition: Best submissions will be considered for awards sponsored by the CCC.

Submission Guidelines

Length: Limited to 4 pages, excluding references.

Format: IEEE 2-column format (see IEEE Templates).

Review Process: Single-blind; submissions must include author names, affiliations, and email addresses.

Submission Link: Submit via CyberChair.

Originality: Papers must not have been published elsewhere or be under consideration at other venues.

Important Dates

Submission Deadline: July 1st, 2026 (AoE)

Notification Date: August 30, 2026

Camera-Ready Deadline: September 15, 2026

Conference Dates: November 12-15, 2026 (in-person presentation required)

Presentation and Awards

At least one author of each accepted paper must complete registration and present the paper in person to be included in the proceedings and program. Remote presentations are not permitted.

With generous support from the Computing Community Consortium (CCC), the best Blue Sky papers will receive awards in the form of travel grants:

• First Prize: $1,000
• Second Prize: $750
• Third Prize: $500

Questions?

For questions regarding submissions, please email the track co-chairs with the subject “ICDM 2026 BlueSky.”

BlueSky Track Co-Chairs:

Jipeng Qiang, Yangzhou University (jpqiang@yzu.edu.cn)

Mengke Li, Shenzhen University (mengkeli@szu.edu.cn)

The Computing Community Consortium has released a new brief, titled The Problem of Problem Selection, to help researchers in the beginning of their journey into SARU research. This new brief is an outcome of the December 2024 visioning workshop Supporting At-Risk Users Through Responsible Computing. The workshop convened 49 researchers and practitioners from academia, industry, and civil society who work directly on issues affecting at-risk users of technology, discussing how to build a more rigorous and coordinated research agenda in this space.

Selecting a Research Topic

This brief is intended to assist all SARU researchers, from the student level to senior researchers. It proposes a thoughtful return to the fundamentals of problem selection, one that reminds us that research isn’t always the right solution to a problem. Rather, the goal of topic selection is to “pick the right problem at the right time.” One must evaluate how research itself will contribute to a desired outcome as well as who stands to benefit from that outcome. 

Many SARU researchers are also drawn to research problems because of a personal connection. This brief highlights the advantages of such connections as well as the challenges they present, reminding researchers that it is necessary to strike a balance between personal investment and disinterested perspective.  

Assessing SARU Research Risks

Beyond these guiding principles, the authors provide a comprehensive reference framework for assessing the risks of potential SARU research questions. Their Socio-Ecological Process Model poses specific questions that encourage researchers to consider the resources required to research their topic from all of the following areas:

• Individuals
• Labs
• Institutions
• Research Communities
• Policy, Funding, and Advocacy

These questions span the entire research life cycle (ideation, study design, study execution, paper-writing, and post-publication publicity). For instance, in the ideation phase, it asks what challenges will be present for individual researchers immersing themselves in the proposed work. And in the publicity phase, it asks what institutional supports are in place to handle potential media attention — both positive and negative.

By considering the risks of SARU research from all angles, researchers can approach their topics both more prepared to mitigate or eliminate them and more certain that their research is ultimately beneficial to affected communities.

Read the SARU Research Problem Selection Brief Here

Tune in to the CCC LinkedIn Showcase Page for updates and more resources like this. Stay connected with CCC for the latest insights, publications, and opportunities to engage by subscribing here.

The Computing Community Consortium (CCC) recently published a new report on how advanced computing, especially artificial intelligence (AI), can extend that impact even further while at the same time contribute to human-in-the-loop computational research. The report, titled Grand Challenges for the Convergence of Computational and Citizen Science Research, assesses the ways that such technology can increase the potential of citizen science, ultimately enhancing scientific capability more broadly. It also lays out the necessary next steps in computing research to make this convergence a reality. 

Increasing Scientific Output

One of the biggest strengths of technology like AI is its ability to process and interpret large amounts of complex data. In addition to generating such datasets, citizen scientists possess the knowledge, creativity, and skills that make up for what such technology cannot do alone. They can enrich models with contextual insights and provide essential edge-case observations that increase the reliability of processed data, making the most of machine learning models while avoiding their downfalls. 

This symbiotic relationship has the potential to support more accurate and timely scientific outputs across fields. For computing researchers specifically, it also creates a unique opportunity to do vital computational research into building trust, access, and transparency into AI systems by embedding them in real-world participatory contexts.

The applications of faster, more reliable data processing are broad. Other areas of scientific research that the report highlights as standing to benefit from convergence include supporting endangered species and protecting them from poachers, accelerating medical research, and improving our knowledge of astrophysics.

A Long-Term Research Roadmap

Further research into advanced computing technologies is required for convergence to truly flourish. Below are some of the foundational investigations required to advance the field, focusing on developing new models, metrics, and frameworks for human-AI interaction, trust, and accountability.

• Incentivize cross-disciplinary and cross-sectoral collaboration: Break down silos by investing in strategies that incentivize collaboration and knowledge sharing between computational and citizen science researchers. Encourage collaboration between academic researchers, federal agencies, industry, and local communities to co-design participatory platforms.
• Develop Human-AI teaming frameworks for public participation: Prototype and study new models for collaboration between AI and citizen scientists, particularly incorporating large language models. Specifically investigate novel systems where complementary roles are driven both by multi-agent decisions and community needs. Explore how these systems balance efficiency with human contextual insight in problems such as task assignment, anomaly detection, and data collection or labeling of large datasets.
• Develop explainable AI for non-expert users: Research novel user interface and data visualization techniques to make AI decision-making transparent and interpretable to the general public, as well as developers or researchers, enabling the general public to gain fluency with AI concepts and better understand how AI is used. Determine to what extent these techniques build trust and enable broader public engagement with AI-driven platforms. 
• Design real-time feedback systems for citizen science: Research systems that enable adaptive, multilingual, and just-in-time digital feedback loops that guide users during data collection and analysis. This includes experimentation with LLMs, AR/VR, edge computing, and mobile-first design. Study these systems for improvement in data quality, user engagement, and learning outcomes.
• Institutionalize evaluation and trust metrics: Fund research on trust diagnostics, engagement dynamics, and societal benefit indicators to guide iterative improvement and accountability.
• Advance participatory AI governance models: Research mechanisms to ensure communities have real governance over how AI is deployed and data are used, including mechanisms for consent, opt-out, accountability, and oversight. Incentivize co-development of toolkits that help projects explain AI behavior to users, including uncertainty visualization, explainable model outputs, and citizen-led model critique workflows. Offer funding (e.g., challenge grants) for co-designed AI tools that are built with community organizations, encouraging broad public participation in scientific research.
• Pilot Human-AI teaming systems across multiple scales and domains: Fund research as well as deployment testbeds at local and global scales in domains where humans and AI collaborate in real time, including disaster response, health, and environmental monitoring, building scalable models for multi-agent systems.

Of course, the success of convergence also depends on the skills and preparedness of the humans involved — including researchers and citizen scientists themselves. Multilingual, psychology-informed, and adaptive systems incorporating AI that train and guide users must be developed and embedded in citizen science projects; such systems will build user capacity and skills while also improving data quality. And in continuing education, it’s essential to build both ML/AI skills and civic literacy, fostering a new generation of convergence-ready scientists, engineers, and public leaders.

Read the Full Report

For a full picture of the impact of large-scale convergence, as well as key recommendations across sectors for how to make it a reality, we encourage all members of the computing community to read the Grand Challenges for the Convergence of Computational and Citizen Science Research report below.

Read the Full Report Here

Tune in to the CCC LinkedIn Showcase Page for updates and more reports like this. Stay connected with the CCC Blog for the latest insights, publications, and opportunities to engage by subscribing here.

Call for Papers: ACM International Conference on Multimodal Interaction

Track: Blue Sky Ideas

Date: October 5-9, 2026

Location: Napoli, Italy

Website: https://icmi.acm.org/2026/call-for-papers-2/

Aims and Scope

The Blue Sky Ideas Track at the ACM International Conference on Multimodal Interaction seeks visionary contributions at the intersection of multimodal artificial intelligence (AI) and social interaction. This track encourages submissions that push the boundaries of conventional research, challenging existing methodologies, proposing novel theories or applications, and presenting high-risk, controversial ideas. Papers should exhibit deep, interdisciplinary insights, robust argumentation, and a synthesis of ideas from various methodologies. 

This year’s theme is “Context and Cultural Awareness for Multimodal Interaction.” A unique aspect of ICMI is its multidisciplinary nature, bringing together research in AI, multimodal data processing, and human-human and human-computer interaction to bridge behavioural understanding with technology, with an eye towards impactful applications that benefit people and society.

Why Submit?

Shape bold, long-term research visions in multimodal AI and human-centered computing

Engage with leaders across the computing field

Gain recognition through CCC and ACM-sponsored awards

Submission Guidelines

Submissions must be no more than 4 pages, excluding references

Please use the double column ACM conference template

Author names and affiliations should be not included

Submit via: https://new.precisionconference.com/submissions/icmi26a

Important Dates

Submission Deadline: Friday, May 8, 2026

Notification of Acceptance: Tuesday, June 23, 2026

Camera-Ready Papers: Friday, July 10, 2026

Presentation Dates: October 6-8, 2026

Publication Policies

As of January 1, 2026, ACM has fully transitioned to Open Access. All ACM publications, including those from ACM-sponsored conferences, will be 100% Open Access. Authors will have two primary options for publishing Open Access articles with ACM: the ACM Open institutional model or by paying Article Processing Charges (APCs). With over 1,800 institutions already part of ACM Open, the majority of ACM-sponsored conference papers will not require APCs from authors or conferences (currently, around 70-75%).

Authors from institutions not participating in ACM Open will need to pay an APC to publish their papers, unless they qualify for a financial or discretionary waiver. To find out whether an APC applies to your article, please consult the list of participating institutions in ACM Open and review the APC Waivers and Discounts Policy. Keep in mind that waivers are rare and are granted based on specific criteria set by ACM.

Understanding that this change could present financial challenges, ACM has approved a temporary subsidy for 2026 to ease the transition and allow more time for institutions to join ACM Open. The subsidy will offer:

$250 APC for ACM/SIG members

$350 for non-members

This represents a 65% discount, funded directly by ACM. Authors are encouraged to help advocate for their institutions to join ACM Open during this transition period.

This temporary subsidized pricing will apply to all conferences scheduled for 2026, including ICMI.

Awards

The first, second, and third place winners will have the opportunity to present their papers at the conference and will be included in the proceedings. Travel grants will be awarded to support their travel to the conference. CCC sponsors awards to honor the first ($1,000), second ($750), and third ($500) place winners in the form of travel grants. In addition, we will further distribute and publicize the three Blue Sky award papers.

Questions?

Contact the organizers at: icmi2026-blueskypapers-chairs@acm.org

Blue Sky Paper Chairs:

Catherine Pelachaud (CNRS – ISIR, Sorbonne University)

Timothy Bickmore (Northeastern University)

To that end, the U.S. Department of Health and Human Services (HHS), together with the Assistant Secretary for Technology Policy and Office of the National Coordinator for Health Information Technology (ASTP/ONC), recently put out a request for information (RFI) on the advancement of AI use in clinical care. It asked what HHS can do to foster public trust and confidence in modern technology solutions, to reduce uncertainty that impedes AI innovation, and to align federal incentives so that AI is deployed in ways that enhance productivity, reduce burden, lower health care costs, and improve health outcomes for patients, caregivers, and communities.

The Computing Community Consortium (CCC), in conjunction with the Computing Research Association – Industry (CRA-I), recently assembled a response to this RFI. In it, computing experts and healthcare professionals come together to present their recommendations for how to make the most of AI advancements in a clinical setting while prioritizing continuing quality of care, patient privacy, and other essential concerns. Below are some of the prevailing themes from the response.

The Potential and Limitations of Clinical AI

One of the biggest barriers facing AI adoption is public hesitancy. Many Americans are reluctant to engage with, or have their healthcare providers engage with, AI tools in a clinical setting because of reliability and privacy concerns. Training AI models with more patient data is likely to improve their output’s usefulness; yet as of now, little assurance exists that that data will remain secure in a HIPAA-compliant way. Because of this, investing in making AI tools secure and trustworthy — and helping the public see this effort — is paramount.

At the same time, studies have found immense potential for AI to enhance clinical care, like assisting in recording patient visits, differential diagnosis, creating plans for therapy and alternatives, and administrative functions like appointment scheduling. Especially in cases where AI can assist providers in avoiding documentation-related burnout, it has incredible potential to enhance the doctor-patient relationship.

This response also highlights ways, moving forward, to improve the overall usefulness and usability of AI tools in healthcare, such as:

• Making sure to consult healthcare professionals on what would best support their work
• Keeping models up-to-date with local, not just aggregate, data
• Avoiding relying on autonomous models
• Investing in industry-academia partnerships across AI and clinical care
• Reforming regulations to improve access to anonymized health data and incentivizing health systems to collaborate with researchers to contextualize that data
• Creating and implementing standard safety evaluation frameworks

For these actions to have maximum impact, it is also crucial to address a growing “AI gap” between clinical settings in urban and rural communities. Devoting resources to supporting clinical care in states in the Established Program to Stimulate Competitive Research (EPSCoR) and Institutional Development Award (IDeA) programs will help to address this divide and promote quality care for all.

Future Research Directions

This RFI response recommends three key areas of research to advance clinical AI use:

• AI-Driven Clinical Decision Support: Developing new end-to-end models that include reasoning and inference to assist in clinical decision-making
• Human-AI Collaboration: Understanding how best to keep humans, including patients, caregivers, clinicians, and more, engaged as key stakeholders in AI tools
• Development and Implementation Trajectory Evaluation Frameworks: Ensuring the information produced by AI models is accurate, relevant, and understandable through the in-context evaluation of clinical AI use.

By investing in these research areas, we can move beyond experimental models and toward systems that are mindful, technologically sound, and seamlessly woven into the complex reality of modern healthcare.

Read the Full Response

For the full scope of CCC, CRA, and CRA-I findings and analysis on advancing AI in clinical care, access the full response below. You can also see more CCC responses to the community here.

Read the Full RFI Response Here

Working alongside CCC staff, the workshop was organized by an inspired team comprising of: Gabrielle Allen (University of Wyoming), Randal Burns (Johns Hopkins University), Sebastian Elbaum (University of Virginia), William Gropp (University of Illinois Urbana-Champaign), Manish Parashar (University of Utah), Nils Aschenbruck (IEEE-CS, Osnabrück University), Terry Benzel (IEEE-CS, University of Southern California), and Rick Kazman (IEEE-CS, University of Hawaii). Their organizing efforts were made possible by the generous support of the National Science Foundation (NSF). 

Fostering Generative Discussion

Through a mix of panels, Q&As, plenary conversation, and small, discussion-heavy breakout sessions, participants had the opportunity to both constantly learn from each other as well as contribute their own expertise to the conversation. With Beyond Code attendees representing a variety of sectors — including academia, industry, government, and NGOs — it led to an enriching experience that brought many perspectives to the table.

Throughout the workshop, participants and speakers touched on a number of timely topics, such as:

• Potential changes in the role of the computer scientist
• Educating future computer scientists to meet this evolving field
• Improving the human-AI relationship
• Making AI output more trustworthy and reliable
• Navigating accountability in the AI landscape

From the most technical conversations about how to actually improve AI output to more meta discussions about its increasing role in our everyday lives, it was clear that scaling AI has more than just technological impacts.

A Roadmap for the Future of AI

These subjects, and more, will be explored fully in a forthcoming CCC-published workshop report. The report is anticipated to be released in mid-2026, synthesizing the many ideas and explorations of the workshop into a concrete set of recommended computing research and policy directions. 

We also invite all members of the computing community to participate in the AI conversation from wherever you are. CCC recently launched the AI Research Ecosystem discussion forum, designed to assess the biggest challenges facing AI research today. We need your input to gauge which challenges are weighing most heavily on the community so that we can better help to address them. 

Stay tuned to the CCC Blog and the CCC LinkedIn Showcase Page for updates and the Beyond Code workshop report’s release. Stay connected with CCC for the latest insights, publications, and opportunities to engage by subscribing here.

The report presents a roadmap for maximizing the potential of citizen science through the contributions of AI — and vice versa — while also demonstrating the broader applications of this union for challenges across ecological, infrastructural, clinical, and other domains. “We are entering a brave new world where we are renegotiating the relationship between humans and machines,” says Lucy Fortson (University of Minnesota), one of the report’s co-chairs. “Investing in human-machine teaming research for citizen science is investing in … accelerating scientific output.”

The report, also co-chaired by Lea Shanley (International Computer Science Institute and GNIES, University of Wisconsin-Madison) and with contributions from lead authors Tanya Berger-Wolf (The Ohio State University), Kevin Crowston (Syracuse University), Corey Jackson (University of Wisconsin-Madison), Saiph Savage (Northeastern University), and Haley Griffin (Computing Research Association), is the culmination of extensive visioning. The findings are most notably informed by discussions at the CCC Grand Challenges for the Convergence of Computational and Citizen Science Research workshop on April 8-9, 2025 in Washington, D.C., as well as two virtual roundtables on the topic. In total, 46 experts across computing research, NGOs, philanthropy, industry, and federal agencies came together to envision “how humans and machines may team up to solve some of the world’s most pressing scientific problems,” articulating specific next steps for making that future a reality.

Key Priorities and Opportunities for Growth

This report centers the immense opportunity that arises when human talent is at the core of emerging technologies. Volunteer citizen scientists are capable of data labeling, analysis, and creativity that machines are simply currently incapable of or may not have the resources to perform. Scaling these citizen science efforts would help close the gap between the sheer volume of data that computational technologies are able to produce and the data-based interpretations scientists can then apply to solving complex questions.

Five strategic priorities are identified for increasing this convergence of computational technology and citizen science:

• Create novel ways for humans and machine learning/AI to interact, enabling multi-agent teams to accelerate scientific discovery while balancing productivity, accuracy, engagement, and the education of participants. 
• Craft better, more responsive feedback loops that connect volunteers, scientists, project teams, and other stakeholders in meaningful ways in order to sustain participation and ensure data reliability.
• Establish trustworthy, transparent, and reliable systems that make volunteers feel respected and included.
• Ensure the security and privacy of data, safeguarded against threats in order to strengthen credibility, improve societal uptake of results, and empower more people to contribute to research. 
• Design and implement an infrastructure that can support large-scale participation and both the human and technological needs that underpin it.

Future Recommendations

These strategic priorities guide detailed recommendations for the future research directions and other actions that support the goal of large-scale convergence. They call on researchers, federal agencies, and industry professionals

• Creating a national infrastructure for convergence that establishes sustained platforms, governance systems, and the physical/cyber architecture required to support scalable, trustworthy, and nationwide convergence efforts.
• Focusing on the foundational scientific and socio-technical investigations required to advance convergence, focusing on developing new models, metrics, and frameworks for human-AI interaction, trust, and accountability.
• Developing the necessary human capital, including the skills, knowledge, and organizational structures, to create, manage, and participate in convergence projects across all sectors.

Read the Full Report

The Grand Challenges for the Convergence of Computational and Citizen Science Research report is available now on the CCC website. It provides a detailed roadmap of not only the full benefits of convergence and human-centered computing, but clear, actionable steps for making it possible. 

We encourage all members of the computing community to read the full report here.

Call for Papers: SEMANTiCS 2026 Blue Sky Ideas Track

Event: SEMANTiCS 2026

Track: Blue Sky Ideas

Date: Tuesday, September 15, 2026 – Thursday, September 17, 2026

Location: Ghent, Belgium

Website: https://2026-eu.semantics.cc/

Aims and Scope

The Blue Sky Ideas Track at SEMANTiCS 2026 seeks visionary ideas addressing long-term challenges and opportunities in the semantic web and knowledge graph research field. We invite submissions of original papers that tackle open questions and unexplored ideas, narrowing in on one of the following foci:

• Solution: Present groundbreaking solution concepts or paradigms to long-term, well-known (conceptual or technical) challenges in the field.
• Problem:Clearly describe and specify a major challenge that, once solved, would have a significant impact on the research community, alike to the “millenium problems” in mathematics.
• Application: Describe a relevant application problem from industry to be solved with knowledge graphs and semantic technologies.
• Cross-disciplinary research: Identify emerging and promising cross-disciplinary research ideas; argue for promising paradigms to be learned and adapted from other research disciplines; discuss the potential impact of knowledge graphs on other research disciplines.
• Societal aspects: Present ideas and visions of contributing to solving societal challenges such as climate change, safeguarding democracy in the digital world, etc.

Some Topics of Interest Include (but are not limited to): 

• Neurosymbolic AI
• Knowledge graphs and language models
• Human-AI interaction and knowledge graphs
• Semantics and distributed ledgers
• Knowledge graphs and robotics
• Hybrid intelligence
• Spatio-temporal and multi-modal knowledge graphs
• Graph foundation models
• Knowledge‑driven data quality & governance
• Trust‑weighted and uncertainty‑aware knowledge graphs
• Cognitive knowledge graphs
• On‑device / edge knowledge processing
• Self‑maintaining / self‑healing knowledge graphs
• Causal knowledge graphs
• The impact of AI agents and agentic systems on knowledge graphs
• Quantum computing and knowledge graphs
• Brain-knowledge graph interfaces

Review Criteria

Submissions will be assessed based on their novelty and quality of argumentation. Specific criteria include the level to which submitted papers: 

• Contribute provocative and out of the box propositions
• Propose high-risk/high-gain ideas
• Differ from the mainstream research
• Address a long-term time horizon till the foreseen adoption by the research community

Submission Guidelines

Submissions must be no more than eight pages, excluding references. Please follow the same formatting guidelines as the SEMANTiCS research track. 

Submit via: https://easychair.org/conferences?conf=semantics2026

Important Dates

Submission Deadline: May 30, 2026

Notification of Acceptance: June 30, 2026

Presentation Date: July 16, 2026

Presentation and Awards

The Program Committee will select three winners to present their work at the SEMANTiCS conference. The specific prizes will then be based on public voting amongst these three submissions. With generous support from CCC, the following prizes will be awarded as travel grants:

First Prize: $1,000

Second Prize: $750

Third Prize: $500

These three finalist papers will be published in the main conference proceedings (IOS Press). All other accepted submissions will be published in CEUR.

Blue Sky Track Chairs

Heiko Paulheim, University of Mannheim, DE

Marta Sabou, Vienna University Economics and Business (WU), AT

Contact email: semantics2026-bluesky@easychair.org

The buzz around artificial intelligence (AI) is undeniable, with daily headlines touting its revolutionary potential. But for AI to truly transform science and society, we need to look beyond the impressive demos and massive models and ensure we achieve the desired impacts in a deliberate, responsible, secure way.

Last week at the AAAS 2026 Annual Conference in Phoenix, Arizona, a panel organized by the Computing Community Consortium (CCC) titled “Shaping the Future of AI’s Impact on Society” captivated a crowded room of researchers and media representatives. Manish Parashar (University of Utah) moderated the panel, and the speakers were Rayid Ghani (Carnegie Mellon University), Carla P. Gomes (Cornell University), and Elham Tabassi (Brookings Institution). Haley Griffin (CCC) and Michela Taufer (University of Tennessee, Knoxville) organized the panel.

Redefining Scale

The theme of this year’s AAAS conference was “Science at Scale.” So what about scaling AI? As Ghani eloquently put it, “Scale means, ‘How do I scale the impact of the work?’ It’s not the model, it’s the impact.” He shared compelling examples from his work as part of the Data Science for Social Good program on preventing homelessness and managing mental health crises, where AI is designed to be proactive and supportive rather than reactive and autonomous. He also emphasized the importance of reusability — while a lot of models are very context-specific and need to be customizable, the ubiquity of many social issues means that a model that benefits one community is likely to help in another. The goal isn’t just a bigger, more complex algorithm, but measurable, positive changes in people’s lives. 

The panelists also challenged the traditional definition of scaling by sharing translational approaches, where research in one domain can be applied to another. As Gomes put it, “It’s easier to get funding for studying birds than for addressing social issues; large language models are powerful across domains.” She explained that models trained to track bird migration, for instance, can also be adapted to predict materials properties or to map poverty. This is especially important when research funding for social issues is limited.

Building Trust Through Robust Evaluation

A significant gap currently exists between AI’s advertised capabilities and its real-world performance. Tabassi, a key architect of the National Institute of Standards and Technology’s AI Risk Management Framework, underscored at the panel that improving trust in AI is fundamentally a socio-technical problem that requires socio-technical solutions. It is a slow, context-specific process that requires transparency about both the limits and capabilities of AI systems. “Trust is hard to build… and it doesn’t scale as fast technology scales,” she noted. She pointed out that current benchmarks for AI success often measure narrow, task-specific capabilities rather than reliable, safe, or privacy-enhancing deployments. 

Tabassi suggested a national infrastructure specifically dedicated to AI evaluation, similar to how the National Artificial Intelligence Research Resource (NAIRR) aims to provide computational data. Gomes emphasized that real-world AI systems often need to be designed with multiple objectives in mind, including social, economic, and environmental considerations, which in turn requires multi-objective Pareto optimization. As an example, she highlighted the strategic planning of hydropower expansion in the Amazon basin, where AI can help meet growing energy needs while reducing adverse impacts on both people and nature.

Addressing Resource Disparities and Systemic Challenges

The conversation also tackled practical hurdles. Academia often lacks the data and computational resources to compete with industry, particularly with foundational models, where the user queries, system responses, and infrastructure are proprietary. This creates an imbalance that hinders independent research and evaluation. Parashar also pointed out the lack of effective models and incentives for private-public partnerships. He highlighted the model being explored by the Responsible AI Initiative in Utah which uses innovation, sandboxes, and regulatory mitigation to catalyze partnerships between academic researchers, entrepreneurs, and government.

Beyond technical concerns, the panelists confronted the ethical quandaries of AI infrastructure itself. A poignant question from the audience highlighted how corporate AI data centers can negatively impact marginalized communities through pollution. The consensus was that while new AI-specific regulations are evolving, existing agencies (such as the Environmental Protection Agency [EPA] and Federal Trade Commission [FTC]) need greater resources and mandates to enforce current regulations in an AI-powered world.

The Path Forward

The panel’s insights offer a roadmap for the future of AI research:

• Prioritize Positive Human Impact: Measure AI success by real-world outcomes and impacts, not just technical model-centric metrics.
• Scientist LLMs: Scientific reasoning is lacking from current LLM systems; to solve a problem, they need to truly understand it, which requires thinking like a scientist.
• Full Stack Governance: AI systems need to have end-to-end governance so that if an error occurs, it is clear which aspect is the cause (e.g., chip designer, model developer, data supplier). The governance needs to withstand robust evaluation.
• Invest in Interdisciplinary Research: Foster collaboration between AI experts and domain scientists to build customized, reasoning-based AI. AI researchers must become pseudo-experts and savvy collaborators in the fields they are helping (e.g., materials science, or ecology, or social sciences).
• Strengthen Evaluation: Develop robust, scientifically valid, transparent, and dynamic AI evaluation infrastructures that assess trustworthiness, reliability, and societal impact in deployed settings.
• Center Impacted Humans: Impacted communities must lead the problem-definition process, and be involved throughout the life of a research project.
• Democratize Access: Provide researchers, particularly in academia, with greater access to data and computing resources to encourage broad innovation and evaluation.
• Empower Regulatory Bodies: Equip existing regulatory agencies to enforce regulations and standards in an evolving AI landscape.
• Promote Partnerships: Develop models, mechanisms, and incentives for public-private partnerships that ensure AI innovations have a responsible impact on science and society at scale.

Get Involved

The AAAS panel was an exciting opportunity for panelists to examine the future of AI with input from community members. It is also part of a broader CCC effort to help create a thriving and responsible AI research ecosystem. We invite all members of the computing community to participate in the AI Research Ecosystem (AIRE) discussion forum, where you can share your opinions about the biggest challenges facing AI research today to shape the research pathways of tomorrow.

The workshop will take place from February 25-26, 2026 at The Westin hotel in San Francisco, CA. There, participants will engage in panels, breakout group discussions, and report-writing alongside fellow experts, ultimately crafting a research roadmap identifying critical advancements to make this goal a reality — all while digging into the broader implications these advancements would have for the field.

Creating a Roadmap for AI’s Future

Beyond Code will center around envisioning the very future of AI software development, not just current breakthroughs. “I want every attendee to walk away with a clearer vision for how AI-powered development tools and the organizations that depend on them must transform to move beyond code…” says workshop organizing committee member Sebastian Elbaum, Professor of Computer Science at the University of Virginia School of Engineering & Applied Science.  “At the end of the second day I hope we have sketched the skeleton of that revolution.”

To do so, Beyond Code includes an agenda of topics and activities designed to make the necessary next steps for the field concrete. It asks participants to work together to assess key challenges and goals in three main areas:

• Methodology: Improving the workflows, deployments, auditing, and runtimes of AI
• Tools: Making AI more secure and trustworthy
• Users: Improving accuracy in capturing user intent, balancing human ergonomics with machine capability, and evaluating how the human developer role shifts as systems are increasingly automated

Following a panel of visionary speakers on the first day, these three pillars will guide multiple sessions of creative, collaborative ideation. But Gabrielle Allen, organizing committee member and Professor at the University of Wyoming School of Computing, says the committee especially hopes to “spark a productive kind of excitement — where keynotes, panels, and working sessions move experts from industry, academia, national labs, and agencies beyond discussion.” 

The final sessions of the workshop will thus focus on wrap-up documentation, informing the creation of a Beyond Code Workshop Report published by CCC in mid-2026 containing key recommendations for the research field, federal agencies, and industry. It’s an important moment for the CCC, whose aim is to catalyze computing experts and advance the field by helping realize these community-driven visions of the future.

The CCC and IEEE Computer Society look forward to welcoming members of the computing community to an energizing and inspiring workshop. To stay informed with any updates as well as the eventual release of the Beyond Code Workshop Report, keep an eye on the CCC blog and LinkedIn Spotlight Page.

—

The Beyond Code workshop is co-sponsored by the IEEE Computer Society.

This material is based upon work supported by the U.S. National Science Foundation (NSF) under Award No. 2300842. This award supports the Computing Community Consortium (CCC), a programmatic committee of the Computing Research Association (CRA).

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.