pure basic research, Bohr’s Quadrant. Agencies such
as NSF and DARPA (among others) often make
investments in these areas. Pasteur’s Quadrant, on
the other hand, represents fundamental research
applied to specific domains. NIH’s interests, for
example, fall into “use-inspired” basic research in the
context of health and well-being. In the context of
the National AI R&D Strategic Plan, Edison’s Quadrant represents the AI research needs of mission agencies that have near-term, agency-specific goals that
are not being addressed by industry. Parts of DoD and
NIJ (among others) fund research of this type.
Aside from the federal agencies, additional input
to the National AI R&D Strategic Plan came from a
thorough review of the open literature on the state of
the field of AI, public discussions at AI-related meetings, an Office of Management and Budget (OMB)
data call across all federal agencies who invest in IT-related R&D, and a request for information (RFI) by
the Office of Science and Technology Policy that
solicited public opinions about how America can best
prepare for an AI future (Felten and Lyons 2016b).
One additional point about the content of the
plan is important to understand. Because R&D in AI
primarily occurs within the discipline of information
technology, the charge for the creation of the plan
was directed to NITRD. Due to the fact that NITRD
oversees (specifically) IT-related R&D coordination
across the federal government, the content of the
plan is exclusively focused on open IT-relevant issues
for AI. Of course, the Artificial Intelligence Task Force
recognized that AI benefits from a variety of perspectives across many other disciplines, including neuroscience, psychology, social and behavioral sciences,
ethics, law, economics, as well as expertise from
across the broad spectrum of application domains,
including agriculture, transportation, and so forth.
Research and development in these other domains is
not included in the strategic plan, however, due to
the IT-centric tasking of the task force. Nevertheless,
a focus on IT-relevant issues still provides a useful
foundation for considering priorities in AI R&D
investments, and their potential benefits across a
wide range of application domains.
Overview of the
AI R&D Strategic Plan
Ultimately the task force defined seven strategic R&D
priorities for AI that are included in the plan. While
the reader is referred to the plan itself for more details
of the primary areas of emphasis, a quick summary is
given here for completeness, taken from the execu-
Strategy 1: Make long-term investments in AI
research. Prioritize investments in the next generation
of AI that will drive discovery and insight and enable
the United States to remain a world leader in AI.
Strategy 2: Develop effective methods for human-
AI collaboration. Rather than replace humans, most AI
systems will collaborate with humans to achieve opti-
mal performance. Research is needed to create effec-
tive interactions between humans and AI systems.
Strategy 3: Understand and address the ethical,
legal, and societal implications of AI. We expect AI
technologies to behave according to the formal and
informal norms to which we hold our fellow humans.
Research is needed to understand the ethical, legal,
and social implications of AI, and to develop methods
for designing AI systems that align with ethical, legal,
and societal goals.
Strategy 4: Ensure the safety and security of AI systems. Before AI systems are in widespread use, assurance is needed that the systems will operate safely and
securely, in a controlled, well-defined, and well-understood manner. Further progress in research is needed
to address this challenge of creating AI systems that
are reliable, dependable, and trustworthy.
Strategy 5: Develop shared public datasets and environments for AI training and testing. The depth, quality, and accuracy of training datasets and resources significantly affect AI performance. Researchers need to
develop high-quality datasets and environments and
enable responsible access to high-quality datasets as
well as to testing and training resources.
Strategy 6: Measure and evaluate AI technologies
through standards and benchmarks. Essential to
advancements in AI are standards, benchmarks, testbeds, and community engagement that guide and
evaluate progress in AI. Additional research is needed
to develop a broad spectrum of evaluative techniques
Strategy 7: Better understand the national AI R&D
workforce needs. Advances in AI will require a strong
community of AI researchers. An improved understanding of current and future R&D workforce
demands in AI is needed to help ensure that sufficient
AI experts are available to address the strategic R&D
areas outlined in this Plan.
A concise organization of these strategies, taken from
the plan, is shown in figure 2. In the bottom row of
this graphic (in dark red) are the cross-cutting R&D
foundations that underpin nearly all areas of AI,
regardless of application. These foundational issues
include ethical, legal, and societal implications,
focusing on fairness, transparency, and accountability by design, as well as ethical AI (strategy 3); safety
and security issues, focusing on explainability and
transparency, building trust, verification and validation, and securing against attacks (strategy 4); a need
for shared data sets and environments for training
and testing, to accelerate the effective development
of AI (strategy 5); and developing standards and
benchmarks to evaluate AI systems (strategy 6).
While not a strict technical challenge, strategy 7
defines the need for a capable AI workforce for developing and using cutting-edge AI approaches. This
strategy also impacts all of AI R&D.
The middle row of figure 2 (in lighter shades of
blue) focuses on the basic areas of R&D that build
upon the cross-cutting foundations of R&D. These