By Matt Hazenbush, Director of Communications and Member Engagement
The Computing Community Consortium (CCC) launched its new Community Chat series with a dynamic discussion on the future of artificial intelligence (AI) research, featuring David Jensen (University of Massachusetts Amherst) and co-authors of the CCC whitepaper Envisioning Possible Futures for AI Research.
Moderated by Peter Harsha (CRA) and introduced by Michela Taufer (University of Tennessee, Knoxville), the session drew participants from academia, industry, and government who explored what might come after today’s era of large language models and foundation models.
Exploring What Comes After Foundation Models
The whitepaper and discussion centered on a central question: What could be the next revolution in AI?
Jensen and the author team — David Danks, Sebastian Elbaum, Mary Lou Maher, William Regli, Matthew Turk, Holly Yanco, Adam Wierman, and Haley Griffin — outlined six possible “AI research futures” that could shape the next generation of artificial intelligence:
- Neuro-symbolic AI – Integrating reasoning and learning for more reliable and interpretable systems.
- Neuromorphic AI – Developing energy-efficient, brain-inspired computing hardware.
- Embodied AI – Enabling systems that perceive, act, and learn within physical environments.
- Multi-agent AI – Creating ecosystems of specialized AI systems that collaborate to solve complex challenges.
- Human-centered AI – Designing AI that complements and enhances human cognition and social intelligence.
- Quantum AI – Applying quantum computing to advance reasoning, modeling, and optimization.
Together, these approaches reflect a shared conviction that advancing AI requires diversity of paradigms, not just scaling of current ones.
“Every time AI has advanced, it’s been because we had the imagination—and the patience—to explore ideas that didn’t fit the dominant paradigm,” said David Jensen, lead author of the whitepaper. “That’s how we got here, and that’s how we’ll get to the next revolution.”
Key Themes from the Discussion
1. Academic Leadership and Federal Support
Panelists agreed that universities and government programs must play a central role in pursuing high-risk, long-horizon AI research.
“It is not immediately clear that U.S.-trained researchers and U.S.-based companies will lead the next AI revolution,” said David Jensen. “That’s why it’s critical that we continue to invest in academic research that explores what comes next.”
They highlighted the role of federal funding in sustaining multidisciplinary institutes capable of advancing neuromorphic, embodied, and quantum research—fields that require long-term collaboration across disciplines.
2. Curiosity and Research Culture
Mary Lou Maher emphasized that future AI breakthroughs will depend on nurturing a culture that rewards curiosity and exploration rather than short-term deliverables.
“Curiosity itself is an incentive,” she said, underscoring that research environments should give students the freedom to follow open-ended questions and to value discovery as much as results.
The panel discussed how fostering curiosity-driven inquiry, particularly among early-career researchers, will be vital to seeding the next paradigm shift.
3. Identifying the Real Research Gaps
William Regli captured this theme succinctly:
“We need to figure out what the hole is, rather than focus on the donut.”
He and others called for researchers to look past today’s successes with large models and identify the unaddressed challenges—such as reasoning, transparency, and energy efficiency—that could define the next wave of progress.
4. Sustainability and Energy Efficiency
The conversation acknowledged that the enormous computational demands of current AI systems are not sustainable. Neuromorphic and embodied AI were highlighted as promising pathways toward lower-power, biologically inspired systems that can maintain performance while reducing environmental costs.
5. Openness, Access, and Democratization
Panelists pointed to open science initiatives like the National Artificial Intelligence Research Resource (NAIRR) as vital to maintaining U.S. leadership and ensuring equitable participation.
Making advanced AI infrastructure, datasets, and educational materials broadly accessible was seen as essential for enabling diverse researchers and institutions to contribute meaningfully.
6. Interdisciplinary and Collaborative Research
Finally, the group stressed that realizing any of these futures will demand tight integration among computer science, neuroscience, social science, and ethics.
Maher and others emphasized that building “socially aware” AI will require collaboration between technical and human-centered disciplines — a point echoed throughout the Q&A.
Watch the Recording and Share the Whitepaper
If you missed the event — or want to revisit the discussion — you can watch the full recording above and download the whitepaper here:
Envisioning Possible Futures for AI Research
We encourage you to share the whitepaper and webinar recording widely with your colleagues, collaborators, and especially with students and early-career researchers who are shaping the future of AI.
Stay connected with the Computing Community Consortium (CCC) to hear about upcoming Community Chats, new white papers, and opportunities to shape future research directions.
This material is based upon work supported by the U.S. National Science Foundation (NSF) under Award Nos. 1734706 and 2300842. These awards support 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.
