Association Affairs

AAAS S&T Policy Forum explores U.S. competitiveness, now and in the future

See allHide authors and affiliations

Science  27 Jul 2018:
Vol. 361, Issue 6400, pp. 374
DOI: 10.1126/science.361.6400.374

Arati Prabhakar of Stanford delivers lecture at AAAS Science & Technology Policy Forum.

PHOTO: MARK FRANCIS JONES/CJVISIONS.COM

The U.S. scientific research and development enterprise stands at a critical juncture that requires the convergence of basic research and commercial applications to more quickly deliver benefits of nextgeneration technological innovation, said Arati Prabhakar in a lecture at the American Association for the Advancement of Science's Forum on Science & Technology Policy.

Prabhakar, a fellow at the Center for Advanced Study in the Behavioral Sciences at Stanford University and former director of both the Defense Advanced Research Projects Agency (DARPA) and the National Institute of Standards and Technology, also called on the R&D community to leverage advances in the social sciences to meet challenges of the coming decades.

Change was a theme that ran through the 43rd annual S&T Policy Forum on 21–22 June at AAAS headquarters in Washington, D.C., focused on U.S. science, innovation, and competitiveness. The forum featured speakers whose careers span public service, academia, and the private sector, including Susan Hockfield, president emerita of the Massachusetts Institute of Technology and chair of the AAAS board, and Peter Lee, corporate vice president of artificial intelligence and research at Microsoft.

“We need new kinds of advances and new ways of working if we are to contribute to our nation for the next handful of decades as we have for the decades just passed,” said Prabhakar, whose career also has traversed Silicon Valley technological start-ups and venture capital firms. “That's what it will take if we are to renew and fulfill R&D's promise of a better future for our country.”

Closing the gap between high-quality, multidisciplinary basic research and the private sector's skillful talent for commercial success calls for novel approaches. Prabhakar pointed to California’s Cyclotron Road and The Engine at MIT as programs that give innovators time and facilities to turn basic research into commercial products.

The convergence of social and behavioral sciences with mathematics, digital technology, and cryptographic methods also is producing technologies. Prabhakar pointed, for instance, to a research collaboration between University of Washington scientists and Enlearn, an educational nonprofit, that is advancing adaptive learning for individual students by linking their responses with visual content. Likewise, a Stanford medical team identified end-of-life care practices able to curb health care costs.

Accelerating R&D development was at the heart of Prabhakar's William D. Carey Lecture, an annual address begun in 1989 to honor former AAAS executive officer Carey by presenting speakers who exemplify his leadership on public policy issues.

On that point, Prabhakar said ethical and societal implications of new technologies are not to be overlooked. “We need to play our part in helping our society make wise choices about the fruits of our labor,” she said.

Microsoft's Lee looked to R&D's future by tracing his journey from disappointing his father, a physics professor, and his mother, a chemistry professor, for pursuing mathematics at the University of Michigan and being drawn “to the crystalline beauty of some mathematical foundations in computer science.”

“It was the stereotypical hard-core art, science Asian upbringing that you might read about in the papers,” Lee deadpanned. He went on to establish a technology office at DARPA and head Carnegie Mellon University's computer science department.

The technology sector's growth, Lee said, is so exponential that it may call for a “fourth paradigm,” to expand science beyond the experimental, theoretical, and computational sciences to data-driven science. One reason: the growth of data centers able to serve, store, and connect to an expanding range of digital tools through networks, a move away from giant computers.

Microsoft's network more than doubles each year, Lee said, noting that it has added more than 250 data centers, including the June launch of an experimental, submerged data center on the seafloor off the Orkney Islands at Scotland's northern point. It is powered through tidal forces, is constructed of recycled materials, and connects to a modest 850 servers.

“We see these sorts of inflection points very infrequently,” said Lee, discussing the growth of hyperscale data computing, the emergence of large-scale machine learning, and the possible inclusion of a new data-driven science paradigm. “We may in fact be in a similar type of inflection point today in the same way that the invention of mass-produced movable type, the Gutenberg press, was an industrial foundation for human knowledge.”

Navigate This Article