Trump's budget treats quantum and AI as engines of the future

The Trump administration's proposal to double government spending on artificial intelligence and boost quantum computing research by the year 2022 signals the administration's desire to keep the U.S. competitive with China, as well as the realization that these technologies will play keys roles in the country's future defense and economy.

If the proposal were to pass — which is a big if — it would represent a major government-funded cash infusion into technologies that have been traditionally developed in the private sector in the U.S., even as it would cut spending on other basic research. Economic rival China, meanwhile, has put tons of state funding into developing AI, leading to worries that the U.S. would inevitably lag behind in the AI arms race. Even if this budget doesn't pass, there's bipartisan support for increased investment in these technologies, and experts suggest the proposal shows how far AI and quantum have come.

Get what matters in tech, in your inbox every morning. Sign up for Source Code.

"I think it reflects a groundswell of interest, and the reason there's a groundswell is because there's good progress," said Dario Gil, director of IBM Research, who sits on President Trump's Council of Advisors on Science and Technology.

Trump's budget includes $210 million for the National Science Foundation (NSF) to invest in quantum information science (QIS), as well as $237 million for the Department of Energy's Office of Science to spend on QIS, and $25 million "to support early stage research for a quantum internet." In total, that's $175 million more for quantum than last year's budget, according to a handout provided by the White House Office of Science and Technology Policy. It also suggests myriad projects for AI research, including $100 million for the Department of Agriculture to test out advanced agricultural systems, and a $850 million budget for the NSF to invest in AI research.

Gil argues that the budget proposal smartly anticipates the role both AI and quantum will soon play across industries — from the pharmaceutical industry to the automotive industry.

White House CTO Michael Kratsios echoed that idea on a call with reporters, saying, "This administration has made the industries of the future a top priority." But it's not just about progress at home; it's about staying competitive. The budget proposal for next year "ensures America maintains its leadership in AI and quantum information science," Kratsios added.

Kratsios was referring to China, which has set up AI centers in major cities, and spent billions on researching various quantum technologies, from securing its communications networks to full-blown quantum computers.

"I think I would say China's probably three to five years behind us right now in terms of quantum computing," Paul Smith-Goodson, the analyst-in-residence on quantum computing at Moor Insights & Strategy, told Protocol. "China's catching up with us rapidly, and they've had a large investment in quantum technologies."

The proposed $25 million to develop a "quantum internet" is one of the most intriguing ideas. The Department of Energy envisions a quantum internet to be a completely secure, entangled connected web for its national labs and private partners. Under Secretary for Science at the Department of Energy Paul Dabbar said that the EU and China have both started research in this field.

"If we don't, others will do it," Dabbar said on the call.

The budget would be a resetting of the scales of investment in core technologies in the U.S. Gil said that over the last 20 years, there has been "significant growth" in investments in research from industry, especially when compared with the 50 years prior to that, where he said that "the lion's share" of investment came from the government.

"People have different lenses: Some are doing it for economic reasons, some are doing it because they've seen the signs on the R&D advancing; some are doing it for security reasons and competitive reasons," Gil said. "But what is common is that no one is questioning that these are foundational areas in science and technology and that the United States needs to be the global leader in them."

Private investment in quantum computing companies and research has skyrocketed in recent years, as research has moved from the theoretical to the practical. In October, researchers from Google announced that they had achieved "quantum supremacy": using a quantum computer to solve a problem that a traditional computer could not. Others in the research community, including IBM, questioned whether the conditions for supremacy had been met by Google's work. Google, IBM, Honeywell, Microsoft, Alibaba, as well as startups like Rigetti, D-Wave, 1QBit and Strangeworks are all vying to commercialize technology that was, until very recently, relegated to basement research labs. (Those startups have raised over $325 million combined to date, according to data compiled by PitchBook.)

The new budget would likely spur research into new applications of quantum computing, such as using quantum computers to design new lithium chemistry for electric-car batteries; figuring out how to pull nitrogen from the air for fertilizers; and new molecular structures for stronger materials.

"If you look at pharma as an example, where the cost of discovery has been going up over time, that's been a challenge as it takes you longer to bring a drug to market," Gil said. "If we have this revolution in computing, we can compress those cycles."

Gil and others Protocol spoke with suggested that much of this research is five to 10 years away from reaching maturity. But that doesn't mean companies need to wait until then to invest. "Value creation is already happening," Gil said.

IBM has dozens of partners working on its quantum platform today, including JPMorgan Chase, which is interested in calculating risk as quickly as possible, Gil said. He compared it to the processors being used in AI research today: "We started working with FPGAs and GPUs in the late 2000s, but we didn't deploy those accelerators in production until 2015 or '16, so we absolutely get the fact that we can have a novel computing architecture that we engage in value creation right now, but deployment may happen in three or five years from now."

Tony Uttley, president of Honeywell Quantum Solutions, told Protocol that an era where problems will be solved on quantum computers far quicker than classical computers is "coming in short order."

But to routinely solve real problems with quantum machines, the foundational hardware that controls the generally unstable qubits needs to be refined, so investments in research are key. Uttley posited that the budget proposal is "continuing to push the capabilities that would need to exist in some of those foundational underlying capabilities and technologies.

"I think it's a great next step, I look at this as a continued interest in emerging technologies of the future," Uttley said.

All of this novel research will likely require a greater workforce with competencies in AI and quantum technologies. Trump's budget proposal also included $50 million for funding in training at community colleges and historically black colleges and universities around AI, Kratsios said.

Uttley agreed that government support for a workforce that's literate in AI and quantum technologies will be needed sooner rather than later. "In particular, quantum needs to have government involvement, not just on a foundational level, making sure that we have a quantum-educated workforce, but also supportive as a customer," he said.