Artificial intelligence (AI), following on the heels of its older sibling RPA (robotic process automation), is no longer waiting to be born, but remains more of a toddler on the Federal IT scene–still learning to walk before trying to run, but bulking up from an appetite for serious Federal government tech interest and investment.
Factors that stand in the way of rapid growth in use of the technology may be fairly said to include inertia, budget, lack of understanding, scarcity of obvious projects, insufficient compute power (legacy data centers), and a dearth of large data sets necessary to leverage the technology.
But a host of Federal IT policy aims and nascent efforts at agencies are providing plenty of push for the AI Age to kick into higher gear, and point to what may become before too long the largest factor in shying away from AI: a lack of imagination.
Turning points on the Federal policy front include:
From the military, to the intelligence community, to civilian agencies, the growth in stated demand for AI projects is impossible to ignore. Just in recent weeks:
Intelligence agency officials ticked off a partial list of AI projects and priorities they’d like to pursue, and identified important long-term benefits from getting into the game including drastically reducing the amount of time analysts spend on lower-level monitoring work, and creating a workforce culture that is more comfortable with taking chances on new technology.
The Pentagon released its AI Strategy that sets a range of goals to support military personnel, protect the country, and devote substantial, unified attention to managing through hundreds of AI project efforts under the umbrella of its Joint Artificial Intelligence Center (JAIC) announced in June 2018. In announcing the AI strategy, the Defense Department made it clear that AI adoption is key to maintaining U.S. military advantage, and that failure to do so will lead to “eroding cohesion among allies and partners [and] reduced access to markets that will contribute to a decline in our prosperity and standard of living.”
And the White House put its stamp of approval on the technology when President Trump issued an AI executive order that focuses on prioritizing Federal government investments in AI-driven projects, and development by Federal agencies of research and development budgets for AI that will support their core missions. The President emphasized that maintaining the U.S.’s leading global position in AI is “of paramount importance to maintaining the economic and national security of the United States.” Moreover, the White House policy directs Federal agencies to spread their resources – data, models, and computing resources – to third parties including academia and industry to further seed the technology and foster public trust in its applications.
Those developments represent just the tip of the larger iceberg of Federal policy supporting further AI use, and in the case of the White House and the Pentagon announcements, what may come to be seen as climactic policy pronouncements that no future administration or agency heads are likely to draw back from.
“The new AI Executive Order has turned attention to AI, but before we can leverage the power of AI to enable our workforce and unlock the value of our data we need to build the compute and data management capabilities,” said Steve Harris, Vice President and General Manager, Dell EMC Federal. “This investment in new infrastructure represents flipping the 80/20 ratio on legacy/modernization spend – these are investments in innovation for enabling human potential,” he said.
Technical, Funding Resources
Outside of last year’s announcement by the Defense Advanced Research Project Agency (DARPA) that it plans to spend $2 billion over multiple years on investments in new and existing projects through its “AI Next” campaign, firm Federal AI spending plans are still hard to spot with precision from public sources such as agency budget reports and congressional appropriations bills.
But it’s inevitable that spending figures will become more visible as top-level administration and agency policy translates down to the get-it-done levels of Federal IT, and success of AI-related projects are likely to give IT leaders and congressional appropriators the impetus to be more public with their plans.
If industry estimates for larger demand for AI and related “cognitive” technologies is any guide, the demand and spending curves on the Federal side are likely to head up. International Data Corp. (IDC) estimated last year that worldwide spending on cognitive and AI systems would hit $19.1 billion in 2018, up 54% from the prior year, and reach more than $52 billion annually by 2021.
On the Federal technology front, while it’s clear that legacy IT infrastructure was not built to handle large-scale AI projects, the other side of that coin means it’s equally true that the inexorable push for Federal agency IT modernization built in part around cloud infrastructure adoption will clear the way, and create the demand, for more AI.
Likewise, the more that Federal IT modernizes, the more that it creates the ability to create vast amounts of data that represents the vital grist for the AI mill.
“Most agencies still have their AI tools deployed in a lab sandbox, which can be limiting,” said Rob Davies, Executive Vice President of Operations, ViON. “In the short-term, agencies need to identify objectives and start pilot programs and build a few real-world wins, before implementing their AI programs more broadly,” he said.
AI considerations also need to be front and center as agencies consider their data center optimization strategies. Many are turning to hybrid environments, and considering data center as-a-service models to accelerate modernization and get the scalability needed for data-intensive emerging technology – RPA, AI, and internet of things (IoT).
And on Capitol Hill, nobody’s really getting in the way of AI save for in areas of broad budget authority that will influence how agencies spend on further adopting the technology, although lawmakers are paying attention to potential downsides of its use.
Late last month, for instance, two House members put forth a non-binding resolution supporting development by Congress of guidelines for ethical development of AI. One of the sponsors, Democratic Rep. Ro Khanna, who represents a portion of California’s Silicon Valley, said the resolution aims to make sure AI “is implemented with thoughtful guidance given the shifting scope of privacy regulations in the digital economy,” while other backers of the resolution said it was important for AI to generate a high degree of trust in general society.
While the resolution represents a stake in the ground on those issues, it’s also worth noting that it’s not a radical proposal given its support by numerous tech-sector giants and the industry groups that represent them.