US government urged to spend $25bn a year on AI or risk falling behind

By on 05/02/2020
CNAS recommends that departments and agencies, including NASA, adopt a phased approach to increasing funding levels to ensure that money spent on AI is done so “effectively and responsibly”. (Photo by Reinhard Link, courtesy flickr).

Congress and the White House should work together to increase annual federal spending on artificial intelligence (AI) research and development to US$25bn (€19.2bn), according to a national security think tank.

This is one of a number of recommendations set out in a report – entitled The American AI Century: A Blueprint for Action – from the Center for a New American Security (CNAS), an influential Washington, DC think tank established in 2007 and funded largely by defence firms. The report also calls for the promotion of international R&D collaboration and for allies to share the cost burden of building new manufacturing facilities for the hardware used in AI technologies, such as semiconductors. If measures are not taken the US risks losing its “technological edge”, the paper says.

The additional investment in AI to an annual total of US$25bn, which the CNAS recommends be in place by 2025, represents a fivefold increase in AI spending earmarked for 2020 but is “affordable… realistic and doable”, according to the report.

“Given the central role AI technologies likely will play in economic growth, geopolitics and global security, and the sharp growth in global spending on AI, this is a modest sum in relative terms,” it says.

It notes that US$25bn would represent less than 19% of the amount requested for all unclassified R&D spending in president Trump’s 2020 budget, and that a large increase in spending in a specific area is not without precedent. Indeed, the 2019 budget requested an US$18.1bn (16.5bn) increase in defence R&D spending over 2018.

“Increases in spending beyond what existing budgets can support is needed to ensure that agencies don’t lag on AI implementation”, the paper says.

The priority, it says, should be to fund high-risk/high-reward basic science research, where private industry has little incentive to invest “but that hold tremendous potential for valuable new knowledge”.

The report recommends that the federal government adopt a phased approach to increasing funding levels, so that the resources are spent effectively and responsibly. The departments and agencies that receive federal R&D monies – including the Department of Defense, the Department of Homeland Security, NASA, and the Environmental Protection Agency – “will require time to plan for expanded research agendas and to formulate relevant metrics to measure progress and effectiveness”, it acknowledges.

International collaboration

CNAS also calls for enhanced international R&D collaboration to align knowledge and experience; improve interoperability; develop norms and principles; enable more efficient standards setting; and to encourage countries to share the cost of certain projects.

“Decades of experience show that joint work with foreign researchers can be done with great benefit and little detriment to US economic and national security,” the report says.

It notes that the UK, France, Japan, Singapore and South Korea have committed hundreds of millions to AI R&D, and that Toronto is a global AI hub. “Each of these locales, and numerous others, are prime candidates for mutually beneficial cooperation,” the report says, adding that “global AI issues – ensuring AI is safe, transparent, explainable, reliable, and resilient – are especially well suited to broad international research cooperation”.

It also calls for public-private partnership and the creation of an international consortium to help ensure “trusted” semiconductor supply chains and to diversify semiconductor manufacturing.

“The US military and intelligence community have special needs for security that go above and beyond what is available in commercial facilities, yet they lack the scale of demand to make a purely government-dedicated [semiconductor] foundry profitable,” the report says, noting that a new plant would cost in the region of US$10-US$20bn (9.1bn-18.2bn). “The DoD and intelligence community should explore novel approaches for public-private partnerships with US companies to build the capability for trusted design, fabrication, packaging and testing.

“Additionally, the United States should establish an international consortium with allies to share the cost burden of building new semiconductor foundries to ensure a trusted and diverse supply chain”. Member nations should include the global leaders in semiconductor manufacturing equipment, which the report’s authors identify as the US, Japan, and the Netherlands.

Other policy recommendations outlined in the report focus on boosting AI talent in the workforce through education and immigration reforms; incentivising private sector AI R&D with tax credits and by easing access to government datasets; establishing norms for appropriate AI use; improving US government readiness for widespread AI adoption, in part by modernising IT processes; and keeping abreast of global AI developments to prevent ‘technology surprise’.

The recommendations build on existing US AI policy, including the Executive Order on Maintaining American Leadership in Artificial Intelligence, introduced last year, and The National Artificial Intelligence R&D Strategic Plan: 2019 update.  

AI ‘Community of Practice’

In a separate development, it was announced last week that the US General Services Administration and the federal chief information officer are to develop a library of AI use cases to which agencies can refer to aid the “thoughtful adoption of AI”.

The AI Community of Practice (CoP) will support the use of AI in federal agencies, including those related to machine learning and deep learning, robotic process automation, robotics, and natural language processing.

Federal employees who are active or interested in AI policy, technology, standards and programmes will be brought together as part of the CoP to facilitate the sharing of best practices and to advance and share tools.

About Mia Hunt

Mia is a journalist and editor with a background in covering commercial property, having been market reports and supplements editor at trade title Property Week and deputy editor of Shopping Centre magazine, now known as Retail Destination. She has also undertaken freelance work for several publications including the preview magazine of international trade show, MAPIC, and TES Global (formerly the Times Educational Supplement) and has produced a white paper on energy efficiency in business for E.ON. Between 2014 and 2016, she was a member of the Revo Customer Experience Committee and an ACE Awards judge. Mia graduated from Kingston University with a first-class degree in journalism and was part of the team that produced The River newspaper, which won Publication of the Year at the Guardian Student Media Awards in 2010.

One Comment

  1. Marky

    07/02/2020 at

    If we don’t concentrate on developing AI fast enough we could destroy the planet before judgment day.

Leave a Reply

Your email address will not be published. Required fields are marked *