AI Safety and Ethics Research Program

AI Safety and Ethics Research Program is sponsored by National Science Foundation (NSF). This program funds interdisciplinary research on the ethical and safety implications of AI technologies.

NSF 22-502: National Artificial Intelligence Research Institutes | NSF - U.S. National Science Foundation Archived funding opportunity This solicitation is archived. Important information for proposers and award recipients All proposals must be submitted in accordance with the requirements specified in the funding opportunity and in the Proposal & Award Policies & Procedures Guide (PAPPG) and its supplements .

All NSF grants and cooperative agreements are subject to the applicable set of NSF award terms and conditions . NSF has updated its research security policies for NSF funded projects. NSF 22-502: National Artificial Intelligence (AI) Research Institutes Accelerating Research, Transforming Society, and Growing the American Workforce Please see related Frequently Asked Questions (FAQs) NSF 22-011 .

Download the solicitation (PDF, 1mb) National Science Foundation Department of Homeland Security, Science & Technology Directorate Institute of Education Sciences, U.S. Department of Education National Institute of Food and Agriculture National Institute of Standards and Technology Office of the Under Secretary of Defense for Research and Engineering Preliminary Proposal Due Date(s) (required) (due by 5 p. m.

submitter's local time): Full Proposal Deadline(s) (due by 5 p. m. submitter's local time): Important Information And Revision Notes This solicitation significantly extends the time and process afforded for the development of proposals.

See the program calendar under Program Description. Preliminary proposals are now required as a method to increase both the quality of eventual full submissions and to reduce the proposers' unnecessary effort in preparation of proposals that are unlikely to be successful for a competition that will result in a small number of awards.

Desiderata for AI Research Institutes, proposal submission instructions, and solicitation-specific review criteria are revised to stress the role and importance of foundational AI research. Desiderata and proposal submission instructions have been revised to emphasize the importance of achieving a whole that is greater than the sum of the parts through both internal synergies and external partner engagement.

New themes for Institute proposals (see Program Description). Agency and Industry partners on this solicitation have changed. Guidelines for the participation of the industry sponsor and its affiliated personnel in proposals to this solicitation apply only to the sponsored theme and are detailed in the program description.

Restrictions apply only to the partner listed in this solicitation. Any proposal submitted in response to this solicitation should be submitted in accordance with the revised NSF Proposal & Award Policies & Procedures Guide (PAPPG) ( NSF 22-1 ), which is effective for proposals submitted, or due, on or after October 4, 2021.

Summary Of Program Requirements National Artificial Intelligence (AI) Research Institutes Artificial Intelligence (AI) has advanced tremendously and today promises personalized healthcare; enhanced national security; improved transportation; and more effective education, to name just a few benefits.

Increased computing power, the availability of large datasets and streaming data, and algorithmic advances in machine learning (ML) have made it possible for AI research and development to create new sectors of the economy and revitalize industries.

Continued advancement, enabled by sustained federal investment and channeled toward issues of national importance, holds the potential for further economic impact and quality-of-life improvements.

This program is a joint government effort between the National Science Foundation (NSF), U.S. Department of Agriculture (USDA) National Institute of Food and Agriculture (NIFA), U.S. Department of Education (ED) Institute of Education Sciences (IES), U.S. Department of Homeland Security (DHS) Science & Technology Directorate (S&T), National Institute of Standards and Technology (NIST), Department of Defense (DOD) Office of the Under Secretary of Defense for Research and Engineering (OUSD (R&E)), and IBM Corporation (IBM).

This program solicitation expands upon the nationwide network established by the first 18 AI Research Institutes to pursue transformational advances in a range of economic sectors, and science and engineering fields.

In this round, the program invites proposals for institutes that have a principal focus in one of the following themes, detailed in the Program Description: Theme 1: Intelligent Agents for Next-Generation Cybersecurity Theme 2: Neural and Cognitive Foundations of Artificial Intelligence Theme 3: AI for Climate-Smart Agriculture and Forestry Theme 4: AI for Decision making Theme 6: AI-Augmented Learning to Expand Education Opportunities and Improve Outcomes Cognizant Program Officer(s): Please note that the following information is current at the time of publishing.

See program website for any updates to the points of contact. AI Research Institutes Program Team, telephone: (703) 292-5111, email: AIInstitutesProgram@nsf. gov Applicable Catalog of Federal Domestic Assistance (CFDA) Number(s): 10.

310 --- USDA-NIFA Agriculture and Food Research Initiative 47. 049 --- Mathematical and Physical Sciences 47. 070 --- Computer and Information Science and Engineering 47.

074 --- Biological Sciences 47. 075 --- Social Behavioral and Economic Sciences 47. 076 --- Education and Human Resources 47.

079 --- Office of International Science and Engineering 47. 083 --- Office of Integrative Activities (OIA) 84. 305 --- Institute of Education Sciences, U.S. Department of Education 97.

108 --- Department of Homeland Security, Science & Technology Directorate Anticipated Type of Award: Cooperative Agreement Estimated Number of Awards: 7 NSF plans to make approximately one Institute award in each of themes 1-5, and one award to each of the two tracks listed in theme 6 as described below.

Anticipated Funding Amount: $140,000,000 Institute awards will be made for between $16,000,000 and $20,000,000 for four to five years ($4,000,000 per year on average). Proposals outside this range may be returned without review. Estimated program budget, number of awards and average award size/duration are subject to the availability of funds.

Who May Submit Proposals: Proposals may only be submitted by the following: Institutions of Higher Education (IHEs) - Two- and four-year IHEs (including community colleges) accredited in, and having a campus located in the US, acting on behalf of their faculty members.

Special Instructions for International Branch Campuses of US IHEs: If the proposal includes funding to be provided to an international branch campus of a US institution of higher education (including through use of subawards and consultant arrangements), the proposer must explain the benefit(s) to the project of performance at the international branch campus, and justify why the project activities cannot be performed at the US campus.

Non-profit, non-academic organizations: Independent museums, observatories, research labs, professional societies and similar organizations in the U.S. associated with educational or research activities. There are no restrictions or limits. Limit on Number of Proposals per Organization: 2 An organization may submit no more than two preliminary proposals to this solicitation as lead institution.

An organization may submit up to two full proposals that correspond to preliminary proposals reviewed under this solicitation. In the event that an organization exceeds these limits, preliminary proposals will be accepted based on earliest date and time of preliminary proposal submission, i.e., the first two preliminary proposals will be accepted, and the remainder will be returned without review.

A full proposal that does not correspond to a preliminary proposal reviewed in this program will be returned without review. Limit on Number of Proposals for Senior Personnel: 1 An individual may be designated as senior personnel (which includes but is not limited to PI or co-PI) on at most one preliminary proposal, and at most one full proposal to this solicitation.

In the event that an individual exceeds this limit, proposals will be accepted based on earliest date and time of submission, i.e., the first compliant preliminary or full proposal will be accepted, and the remainder will be returned without review. Proposal Preparation and Submission Instructions A. Proposal Preparation Instructions Letters of Intent: Not required Preliminary Proposals: Submission of Preliminary Proposals is required.

Please see the full text of this solicitation for further information. Full Proposals submitted via FastLane: NSF Proposal and Award Policies and Procedures Guide (PAPPG) guidelines apply. The complete text of the PAPPG is available electronically on the NSF website at: https://www.

nsf. gov/publications/pub_summ. jsp?

ods_key=pappg . Full Proposals submitted via Grants. gov: NSF Grants.

gov Application Guide: A Guide for the Preparation and Submission of NSF Applications via Grants. gov guidelines apply (Note: The NSF Grants. gov Application Guide is available on the Grants.

gov website and on the NSF website at: https://www. nsf. gov/publications/pub_summ.

jsp? ods_key=grantsgovguide ). Cost Sharing Requirements: Inclusion of voluntary committed cost sharing is prohibited.

Indirect Cost (F&A) Limitations: The following instructions apply to awards made by USDA-NIFA: For awards made by USDA-NIFA under this solicitation, Section 1462(a) and (c) of the National Agricultural Research, Extension, and Teaching Policy Act of 1977 (NARETPA) (7 U.S.C. 3310) limits indirect costs for the overall award to 30 percent of Total Federal Funds Awarded (TFFA) under a research, education, or extension grant.

The maximum indirect cost rate allowed under the award is determined by calculating the amount of indirect costs using: 1) the sum of an institution's negotiated indirect cost rate and the indirect cost rate charged by subawardees, if any; or 2) 30 percent of TFFA (TFFA = Field K. , Total Costs and Fee, on SF-424 R&R Budget).

The maximum allowable indirect cost rate under the award, including the indirect costs charged by the subawardee(s), if any, is the lesser of the two rates. If the results of 1), is the lesser of the two, the grant recipient is allowed to charge the negotiated indirect cost rate on the prime award and the subaward(s), if any. Any subawards would be subject to the subawardee's negotiated indirect cost rate.

The subawardee may charge its negotiated indirect cost rate on its portion of the award, provided the sum of the indirect cost rate charged under the award by the prime awardee and the subawardee(s) does not exceed 30 percent of the TFFA. If the result of 2), is the lesser of the two, then the maximum indirect cost rate allowed for the overall award, including any subaward(s), is limited to 30 percent of the TFFA.

That is, the indirect costs of the prime awardee plus the sum of the indirect costs charged by the subawardee(s), if any, may not exceed 30 percent of the TFFA. In the event of an award, the prime awardee is responsible for ensuring the maximum indirect cost allowed for the award is not exceeded when combining indirect costs for the Federal portion (i.e., prime and subawardee(s)) and any applicable cost-sharing (see 7 CFR 3430. 52(b)).

Amounts exceeding the maximum allowable indirect cost is considered unallowable and will be handled accordingly. See 2 CFR 200. 408 and 2 CFR 200.

410. Other Budgetary Limitations: Other budgetary limitations apply. Please see the full text of this solicitation for further information.

Preliminary Proposal Due Date(s) (required) (due by 5 p. m. submitter's local time): Full Proposal Deadline(s) (due by 5 p.

m. submitter's local time): Proposal Review Information Criteria National Science Board approved criteria. Additional merit review criteria apply.

Please see the full text of this solicitation for further information. Award Administration Information Additional award conditions apply. Please see the full text of this solicitation for further information.

Additional reporting requirements apply. Please see the full text of this solicitation for further information. AI is advancing rapidly, enabled and significantly fueled by federally-funded basic research.

Increasingly sophisticated and integrated approaches for AI systems appear in applications across all sectors of the economy, and new challenges emerge for advancing, applying, and governing these promising technologies. AI holds the potential to transform lives across our Nation through increased economic prosperity, improved educational opportunities and quality of life, and enhanced security.

At the same time, the potential capabilities and complexities of AI, combined with the wealth of interactions with human users and the environment, make it critically important to further advance our understanding of AI, including aspects of transparency, security, and control.

Among federal research investments, institute-scale activities enable multidisciplinary, multi-stakeholder teams to focus on larger-scale, longer-time horizon challenges in both foundational and use-inspired AI research, and development of the future AI workforce, as well as addressing some of society's grand challenges.

National AI Research Institutes will serve as national nexus points for collaborative efforts spanning institutions of higher education, federal agencies, industry, and nonprofits/foundations in such areas. They should also accelerate the transition of AI innovations into many economic sectors, and nurture and grow the next generation of talent.

A long-term, substantive, and highly visible investment in AI research, infrastructure, and workforce development will realize the potential of, and enable the U.S. to maintain global leadership in, AI. AI enables computers and other automated systems to perform tasks that have historically required human cognition and human decision-making abilities.

Research in AI is therefore concerned with the understanding of the mechanisms underlying thought and intelligent behavior and their implementation in machines.

The full AI endeavor is inherently multidisciplinary, encompassing the research necessary to understand and develop systems that perceive, learn, reason, communicate, and act in the world; exhibit flexibility, resourcefulness, creativity, real-time responsiveness, and long-term reflection; use a variety of representation or reasoning approaches; and demonstrate competence in complex environments and social contexts.

What is sometimes referred to as "core AI" research addresses, in general, the theory and methods that give rise to these target abilities and their implementation in machines. It includes research in all matters of learning, abstraction, and inference required for intelligent behavior as well as general architectures for intelligence, integrated intelligent agents, and multi-agent systems.

Machine learning, that is, methods for solving tasks by generalizing from data, has made great advances in recent years through the combination of new algorithms, increases in computing power, and the growing availability of data.

Machine learning does not, however, encompass all of core AI; that also includes research on knowledge representation, logical and probabilistic reasoning, planning, search, constraint satisfaction, and optimization. In some lines of AI research, computational models and mechanisms of intelligence draw direct inspiration from living systems.

Biologically-inspired computing draws from connectionism, behavior, and emergence in living systems to inform algorithm and system design. Computational neuroscience contributes models based on theory and analysis of computational processes in the nervous system.

Behavioral and cognitive science informs much of the motivation and design of systems seeking to implement behavior typical of human perceptual, motor, and cognitive processes and their interactions. Perception and communication are critical capabilities associated with intelligent behavior. Where AI is concerned, the field of computer vision studies methods that enable systems to sense and reason about the visual world.

Human language technologies (also known as "natural language processing" and "natural language understanding") research enables intelligent systems to analyze, produce, translate, and respond to human text and speech. Intelligent systems may be able to act upon the world through embodiment. Robotics is closely aligned with but not identical to embodied AI.

While an embodied AI may be a robot, this solicitation does not include in the scope of this definition tele-operated robots or industrial robots that simply repeat programmed patterns of motion.

As intelligent systems amplify humans' capabilities to accomplish individual and collective goals, research is needed to assess the benefits, effects, and risks of AI-enabled computing systems; and to understand how human, technical, and contextual aspects of systems interact to shape those effects.

Relevant research areas therefore include consideration of explainable and trustworthy AI; validation of AI-enabled systems; AI safety, security, and privacy; and the role of emotion and affect in the design and perception of increasingly sophisticated machine intelligence.

Research in AI also encompasses novel software and hardware architectures, as well as methods for carrying out AI algorithms on a variety of computing systems and platforms, including those that operate under additional constraints such as time (e.g., real-time) or energy, or those targeting specific application classes or use cases.

Developing hardware further optimized for AI and ML algorithms or hardware offers the potential for even higher levels of performance. The above definition of AI and its principal disciplines establishes the scope of this National AI Research Institutes program. I.

B. Foundational and Use-Inspired AI Research Research in foundational AI seeks to develop theory and methods that are independent of any particular domain of application. Use-inspired AI research refers to basic research that has use for society in mind.

Use-inspired research seeks new methods and understanding in AI by situating the research in a domain of application to simultaneously inform progress in AI and solve particular use cases. As an example, foundational research in machine learning gave rise to breakthroughs in deep neural networks motivated by performance in controlled contexts like character recognition.

Later, use-inspired research in the intersection of machine learning and linguistics led to the development of recurrent neural networks in AI while also revolutionizing language modeling for speech and text processing.

We use the phrase "use-inspired" rather than "applied" to emphasize that this solicitation seeks to support work that goes beyond merely applying known techniques and adds new knowledge and understanding in both foundational AI and use-inspired domains.

Ideally there is a virtuous cycle between foundational and use-inspired research, where foundational results provide a starting point for use-inspired research, and the results from use-inspired research are generalized and made foundational. AI has advanced tremendously and today promises personalized healthcare; enhanced national security; improved transportation; and more effective education, to name just a few benefits.

Increased computing power, the availability of large datasets and streaming data, and algorithmic advances in ML have made it possible for AI development to create new sectors of the economy and revitalize industries. Continued advancement, enabled by sustained federal investment and channeled toward issues of national importance, holds the potential for further economic impact and quality-of-life improvements.

Building upon the network established by the first 18 AI Research Institutes, this National AI Research Institutes program solicitation will fund Institutes comprised of scientists, engineers, and educators united by a common focus on advancing the research frontiers in AI. The program seeks to build a broader nationwide network to pursue transformational advances in a range of economic sectors, and science and engineering fields.

AI Research Institutes will have as their primary focus the advancement of multidisciplinary, multi-stakeholder research on larger-scale, longer-time-horizon challenges in AI research than are supported in typical research grants.

They will accelerate the development of transformational technologies by grounding that research in critical application sectors that can serve as motivation for foundational research advances and provide opportunities for the effective fielding of AI-powered innovation. II. A.

AI Research Institutes Scope The vision of the National AI Research Institutes program is broad and ambitious. It is expected that each AI Research Institute will pursue this vision in ways that are uniquely suited to its selected research focus, facilities, collaborations, and other unique circumstances.

Proposers are encouraged to convey the unique qualities of the proposed Institute, while addressing the following desiderata common to all AI Research Institutes proposed to this program: AI Research Institutes advance foundational AI research that will have broad and lasting impact, contributing new knowledge or methods toward understanding of the mechanisms underlying thought and intelligent behavior and their implementation in machines (see the definition of AI specified above).

Institutes aimed at advancing established AI lines of research should demonstrate the potential to radically advance these areas beyond the state of the art. Institutes might also address new foundational AI research priorities that arise from rapid advances in AI and the increasing ubiquity of AI-enabled technology.

Institute proposals that do not describe a clear plan to achieve ambitious advances in foundational AI research are not likely to be responsive to this solicitation . AI Research Institutes conduct use-inspired research that both informs foundational AI advances and drives innovations in related sectors of science and engineering, segments of the economy, or societal needs.

Effective use-inspired research achieves synergy among a group of researchers to enable transformative advances in AI, related sectors, and the interfaces between these areas.

This dimension of an AI Research Institute will feature clear and compelling goals to advance AI and to accelerate the fielding of AI-powered innovation; it also enhances the transfer of knowledge through the meaningful exchange of scientific and technical information with external stakeholders such as industrial partners, public policy makers, or international organizations, as well as with the broader scientific and educational community.

Through use-inspired research, Institutes have the potential to create and share new community infrastructure, including data and software, to further research, promote reproducibility, and support education.

It is critical that proposals clearly specify how the use-inspired context for Institute research reveals the opportunities for foundational AI advances and how those foundational AI advances in turn contribute to the related sectors that define the use-inspired context. AI Research Institutes actively build the next generation of talent for a diverse, well-trained workforce.

Specifically, AI Research Institutes should leverage the visionary nature of their research foci to drive new and innovative education and development tailored toward e.g. undergraduates, graduate students, and post-doctoral researchers, as well as through community colleges and skilled technical workforce training and other opportunities as appropriate that advance knowledge and education of AI, including public understanding of AI.

This could include innovative pedagogy and instructional materials, advanced learning technologies, project-driven training, cross-disciplinary and collaborative research, industry partnerships, and new career pathways.

Institutes should offer broad, deep, and diverse experiences to build the next generation of the AI workforce, with a focus on broadening participation among the full range of groups currently under-represented in science and engineering. AI Research Institutes should maximize their unique position to grow the next generation of talent that will provide new discoveries and leadership.

AI Research Institutes are coherent multidisciplinary groups of scientists, engineers and educators appropriate for a large-scale, long-term research agenda for the advancement of AI and the fielding of AI-powered innovation in application sectors of national importance.

The multidisciplinary nature of these Institutes will catalyze foresight and adaptability beyond what is possible in single research projects; further, the individual projects that an Institute carries out should meaningfully integrate into fundamental contributions beyond the sum of the individual projects. Each Institute will be comprised of multiple organizations working together to create significant new research capabilities .

NSF and partner organizations seek to grow the network of National AI Research Institutes in lead organizations distributed throughout the country to grow new centers of AI leadership and leveraging existing centers of excellence as appropriate.

Institutes are strongly encouraged to include organizations that can directly contribute to NSF's commitment to broadening participation by engaging a diverse, globally engaged research community, integrating research with education and building capacity, and expanding efforts to broaden participation from underrepresented groups and diverse institutions across all geographical regions.

Participants should be meaningfully integrated into a diverse Institute that is more than just the sum of the parts. Each Institute will have a lead PI with demonstrated vision, experience, and capacity to manage a complex, multi-faceted, and innovative enterprise that integrates research, education, broadening participation, and knowledge transfer.

Each Institute will also be staffed with a Managing Director or Project Manager (distinct from the lead PI) and a suitable Management Team to oversee the operations of the Institute. An External Advisory Board is required for all AI Research Institutes. (Potential Advisory Board members should not be approached or identified until the Institute is funded.)

AI Research Institutes are nexus points for collaborative efforts .

The "nexus point" function in this program is not a mere state of being, but rather an active set of priorities, programs, mechanisms, etc., whereby an AI Research Institute pursues the continuing growth of collaborations with external partners to bring together people, ideas, problems, and technical approaches for maximum impact beyond the members and the boundaries of the Institute itself.

As nexus points, Institutes have the potential to continue to connect with new partners with the best teams and approaches from institutions of higher education, federal agencies, industry, nonprofits/foundations, centers/institutes, and national networks.

As nexus points, Institutes promote organizational collaborations and linkages within and between campuses, schools, and the world beyond, and further the Institute's mission to broaden participation in research, education, and knowledge transfer activities through a network of partners and affiliates. In this round of Institutes, proposals are being solicited in the following high-priority areas.

Submissions MUST have as a principal focus one of the following themes. Theme 1: Intelligent Agents for Next-Generation Cybersecurity The U.S. invests significantly in cybersecurity R&D each year, defending government agencies, companies, critical infrastructure, and educational and health organizations from increasingly diverse and sophisticated threats to computing and cyberinfrastructure.

The hazards of cyber-attacks include loss of intellectual property or even ability to function, privacy risks and misuse of personal information for fraud or blackmail, and the fabrication or spread of disinformation.

Significant advances have been made in harnessing recent advances in AI to monitor, predict, prevent, and respond to specific cybersecurity-related vulnerabilities; for example, by recognizing suspicious patterns in user activities, system traces, and network traffic. As cuber defenses and AI technologies become more sophisticated, however, so do cyber-attacks.

Next generation cybersecurity must go beyond local, individual vulnerabilities to safeguard against complex attacks involving multiple sophisticated and intelligent adversaries carrying out coordinated strikes on multiple systems. These threats pose grave problems for both human security teams and the purely data-driven machine-learning based tools they use because of their scale, novelty, and complexity.

The goal of this Institute is to prepare us for a future where both defenders and attackers increasingly use AI-powered tools as they pursue cybersecurity-related goals. Research at the Institute will deepen knowledge and capabilities in both AI and cybersecurity through transformational advances in agent-based AI systems driven by cybersecurity needs.

In this theme we use the term "agent" to refer to an AI system that frames the overall challenge as one of appropriately autonomous reasoning and action in a complex multi-agent environment.

The Institute will develop methods for modeling the actions, beliefs, and goals of the machine and human agents involved in cyber-attacks and cuber defenses, improving the ability of defenders to deter, protect, detect, and respond to cybersecurity risks.

The research should also account for the limitations and uncertainties of these agents, ensuring that they operate responsibly and in concert with human supervisory control in security operations.

A non-exhaustive list of potential research topics at this Institute includes: AI-enabled analytics across multiple kinds of data for modeling cybersecurity threats, including, but not limited to: natural language intelligence reports from threat reporting registries; dark web chatter; network, process, and access log data; and sensitive data controlled and protected by defenders who need to cooperate while maintaining privacy and security of their own data.

Plan recognition, plan generation, and flexible plan execution for threat deterrence, detection, and response. This might include intelligent prioritization of network traffic and information, automated synthesis of code patches and system configurations, active defenses and strategic counterattacks, developing and deploying counter narratives to misinformation, and other capabilities that reduce cybersecurity risks.

Tools for reasoning about uncertainty in a system's evaluations, managing mistakes and mitigating risks around them, and acting with appropriate levels of autonomy in coordination with existing security responsibility and operations structures. Principled methods for interaction and coordination between humans and AI agents working together in cybersecurity teams.

The methods should take into account the differing strengths and limitations of human experts and AI components, including adapting to best leverage the expertise of particular members on the team and supporting the transfer of the expertise from humans to AI agents. Defenses against adversarial methods that sophisticated attackers would deploy against the proposed foundational AI advances.

For the topics above, these might include inducing incorrect models of human objectives, capabilities, and methods; poisoning or evading machine learning-based components of larger agent-based AI systems; distorting data collection and analysis methods in threat intelligence systems; inducing attacker-friendly or defense-costly response plans; misleading system evaluations of confidence and appropriate autonomy; and interfering with defender coordination.

Methods for recognizing multi-step, multi-agent attack plans from incomplete and uncertain information. This could include the ability to discover new attacks from first principles, and the provenance and evolution of new attacks from existing ones. Game-theoretic models of cybersecurity, in which agents may deploy strategies designed to mislead their opponents.

These strategies may also need to account for coordination and competition between individual attackers and defenders, and the networks of other agents they can control or influence. The Department of Homeland Security Science and Technology Directorate (DHS S&T) and IBM Corporation are providing partial support for this institute theme.

Theme 2: Neural and Cognitive Foundations of Artificial Intelligence Advances in our understanding of neural, biological, and cognitive processes provide a rich set of models and mechanisms for guiding the development of AI to abilities and levels of performance comparable to that of humans.

Deep integration of these advances with theoretical and algorithmic advances in AI have the potential to transform AI conceptualization and implementation (from algorithms to hardware). Likewise, in neuroscience and cognitive science, AI is beginning to transform data analysis and model discovery but has yet to be fully integrated into the theory and foundations of the fields.

The purpose of an Institute in the Neural and Cognitive Foundations of AI is to unify and jointly raise the expectations of these fields, taking full advantage of recent developments across disciplines and new opportunities to develop shared models, abstractions, and common frameworks for research.

In this way, research at the Institute will advance our understanding of intelligence, both in nature and in engineered systems, and build upon these advances to explain biological intelligence and design next-generation AI.

Proposals to this theme are encouraged to convey a research strategy that is visionary, but also timely and appropriately scoped, as the breadth of potential advances in this theme is beyond the capacity of a single Institute to address.

Whatever its chosen focus, research at the Institute is expected to leverage and extend the current state of the art across levels of abstraction in understanding and modeling intelligence, and to explicate the relationships between biological and artificial intelligence.

The topics listed below are illustrative of the range of research questions appropriate for this theme: Neural and cognitive theories that inform the design of AI systems with the potential for active and continuous learning across diverse tasks, dynamic environments, and self-motivated exploration.

This may include theories of brain development, plasticity, and regulatory processes (e.g., sleep, dreaming, and memory consolidation) for their potential role and impact on artificial neural networks. How insights and mathematical concepts that enable AI systems to solve complex perceptual, reasoning, and