White House Summit on Artificial Intelligence in Government WHITE HOUSE SUMMIT ON ARTIFICIAL INTELLIGENCE IN GOVERNMENT
Governments around the world are getting increasingly more serious about artificial intelligence. Not only are governments investing their time and energy at the most strategic levels, but so too are they increasing the expenditures on AI investments. However, simply throwing money at AI research and development is not enough to guarantee that countries will see strategic returns from their investments. While the United States has long been at the forefront in AI development and research, it’s not certain that the US will retain any sort of leadership position in the future. China, the United Kingdom, Germany, France, other European Union countries, South Korea, Australia, Saudi Arabia, and many other nations are investing heavily to achieve a foothold in the world’s race to AI domination.
Recently, the White House held a strategic meeting with major government, industry, and academic leaders focusing on short-term challenges and needs to advance AI in the United States. Chaired by Michael Kratsios, Chief Technology Officer of the United States, along with 175 leaders and experts from government, industry, and academia, the meeting aimed to focus the government’s attention on technical, talent, and workforce issues that are challenging government AI adoption. The government hopes to use summits like this one as well as industry events and meetings such as the growing AI in Government meetup as a way to encourage more dialogue and cross-pollination of ideas and experience among government agencies, industry, and researchers. (Disclosure: Cognilytica is an organizer of the AI in Government meetup)
US Government’s Commitment to AI
The Trump Administration has long seen AI as a strategic initiative. So much so that they launched the American AI Initiative, as a “whole-of-government national strategy for U.S. leadership in AI." Core to this initiative are investments in AI research and development, encouraging the spread of data and resources to support AI initiatives, setting governance standards, as well as growing the AI workforce in government and industry. As part of this initiative, AI.gov was launched as a way to centralize AI efforts, share AI activities, and push forward initiatives that can help spur adoption and growth of AI across government agencies and departments.
As a follow-up to all this activity, the government recently announced $973 Million in funding for AI as part of its 2020 budget proposal, all of which is going to non-defense initiatives. This is a major addition to government sponsorship of AI efforts but is a drop in the bucket when gauged against worldwide investment in AI. Indeed, just one startup alone in China has raised more than $1 Billion for its facial recognition technology. Other companies in the RPA space have collectively raised almost $1 Billion. So while $973M is indeed substantial, it is almost insignificant when compared with venture capital and other investments being made in the industry - much of it outside US borders.
The AI Skills & Knowledge Gap Widening
The White House Summit identified a key challenge to AI leadership is the dearth of skills and experience of AI in the government. In an environment where experienced people in AI can easily command mid-to-high six figure salaries, it’s hard for the government to compete for this talent. The shortage of skilled data scientists, data engineers, machine learning developers and architects of all types is putting a significant crimp on the government’s ability to achieve its AI objectives making it difficult for the government to compete with private industry in this area.
One approach to solving this challenge is to engage third-party experts who can “upskill” existing government workers to achieve fundamental, necessary AI knowledge and capabilities. While training traditional IT talent to become machine-learning experts might be a heavy lift, the AI industry is moving to give non-researchers the same sort of capabilities with data science and machine-learning tools as would have been required by experts just a decade ago. Armed with best practices know-how and adequate training, even mid-level government workers can quickly become adept at implementing any of the Seven Patterns of AI that’s being realized as critical to making AI implementations work in the real world.
Likewise, the White House Summit quickly identified that the lack of best practices knowledge and experience is hampering widespread adoption of AI within agencies and departments. Vendors are not sharing their internal methodologies, which are not meant for government adoption in any case. Likewise, major government contractors see their methodologies as a key differentiator, which limits broad adoption of best practices for AI in government. Emerging methodologies such as the Cognitive Project Management for AI (CPMAI) methodology has been gaining increasing traction as a vendor-neutral approach to combine the best of Agile Methodologies with established data-centric methodologies for project management. While there are other methodologies around, the government must select one or two such approaches, expand the knowledge of these best-practices approaches within government through training, and mandate that AI projects utilize an accepted methodology to guarantee project success. Otherwise, we have the "wild west" of money going out the door to fund AI projects without any guarantee of successful project return.
Takeaways and Next Steps
The White House summit identified several key takeaways to ensure continued momentum for AI in government. They realize that there needs to be a better understanding of best practices, as outlined above. The main means to accomplish this is through increased sharing and collaboration between government, industry, and academia, while also leveraging existing AI project successes realized in different corners of the government. At the summit, attendees discussed which AI patterns are easiest to achieve, and can result in the shortest-term return of investment. While this approach of replicating success is a good one, it tends to lead to circular thinking. The government will only implement those things that have already been successful in the government, missing out on what's happening outside of government and reducing risk-taking, while also rewarding entrenched incumbents in the technology space. The key to bringing in this outside perspective is engaging with others who can bring both industry and government success cases and examples from a wider array of AI use cases and applications.
Another way of achieving the above objective of increased sharing is the idea of the creation of an AI Center of Excellence (COE). At the summit, attendees discussed a model for an AI COE and how it would operate. There are already some examples of this within the government, such as the Joint AI Center (JAIC), as well as emerging academic-government-industry collaboration efforts that can serve both government as well as industry needs. The idea would be that an AI COE would help foster partnerships, investment, and scaling of AI efforts and expand cross-organizational communication of best practices for AI adoption.
Engagement of cross-government and academic groups such as ATARC will also help to widen adoption by increasing sharing as well as movement to standardize and otherwise motivate the setting of methods that help to establish fundamental methods for doing AI right in the government. Likewise, while the group of 175 government officials, industry vendors, and academic organizations was certainly representative of opinions and implementations across a wide range of sectors, it was missing the critical group of industry analysts, AI training and talent development organizations, and others that are more attuned to the use cases, best practices, and organizational growth approaches that were identified as key gaps at the summit.
The summit also uncovered the major challenge of organizational change management. As AI technologies are brought into formerly paper and human-bound processes, governmental change managers need to find ways to build trust and gain acceptance from day-to-day government workers. Indeed, "organizational antibodies" can easily form to resist the adoption of AI technologies and sabotage AI efforts. One way to avoid this is to involve those day-to-day workers in the AI initiatives and to share success stories that show how their jobs are not at risk, but rather can grow and expand as AI projects eliminate portions of their jobs that are repetitive, dull, error-prone, or otherwise not conducive to the overall mission.
While it’s clear that the US government is focused on making AI a priority and a strategic initiative, the summit, with its limited audience and limited visibility, and the government's significant, but limited funding, will only be able to achieve its overall objectives if it can successfully engage with the broader industry in the US. If government can successfully tap the tens of billions of dollars available from the venture community, engage with best practices and thought leadership institutions that are also heavily engaged in industry, and share its needs beyond a small group of well-connected individuals, then the US will indeed be able to pursue and maintain its goal of being a leader in the AI ecosystem.
Ronald Schmelzer, columnist, is senior analyst and founder of the Artificial Intelligence-focused analyst and advisory firm Cognilytica, and is also the host of the AI Today podcast, SXSW Innovation Awards Judge, founder and operator of TechBreakfast demo format events, and an expert in AI, Machine Learning, Enterprise Architecture, venture capital, startup and entrepreneurial ecosystems, and more. Prior to founding Cognilytica, Ron founded and ran ZapThink, an industry analyst firm focused on Service-Oriented Architecture (SOA), Cloud Computing, Web Services, XML, & Enterprise Architecture, which was acquired by Dovel Technologies in August 2011.
Ron is a Parallel Entrepreneur, having started and sold a number of successful companies. The companies Ron has started and run have collectively employed hundreds of people, raised over $60M in Venture funding and exits in the millions. Ron was founder and chief organizer of TechBreakfast – the largest monthly morning tech meetup in the nation with over 50,000 members and 3000+ attendees at the monthly events across the US including Baltimore, DC, NY, Boston, Austin, Silicon Valley, Philadelphia, Raleigh and more.
He was also founder and CEO at Bizelo, a SaaS company focused on small business apps, and was Founder and CTO of ChannelWave, an enterprise software company which raised $60M+ in VC funding and subsequently acquired by Click Commerce, a publicly traded company. Ron founded and was CEO of VirtuMall and VirtuFlex from 1994-1998, and hired the CEO before it merged with ChannelWave.
Ron is a well-known expert in IT, Software-as-a-Service (SaaS), XML, Web Services, and Service-Oriented Architecture (SOA). He is well regarded as a startup marketing & sales adviser, and is currently mentor & investor in the TechStars seed stage investment program, where he has been involved since 2009. In addition, he is a judge of SXSW Interactive Awards and served on standards bodies such as RosettaNet, UDDI, and ebXML.
Ron is the lead author of XML And Web Services Unleashed (SAMS 2002) and co-author of Service-Orient or Be Doomed (Wiley 2006) with Jason Bloomberg. Ron received a B.S. degree in Computer Science and Engineering from Massachusetts Institute of Technology (MIT) and MBA from Johns Hopkins University.