By: Chris Katsantonis

On February 11, 2019, President Trump signed an Executive Order (hereinafter “Order”) on “Maintaining American Leadership on Artificial Intelligence.”[1] The Order is not much of a surprise considering China’s recent push to become the leader in Artificial Intelligence (“AI”).[2]  The Order aims to encourage advancements in technology, international and industry collaboration with foreign entities, while ensuring economic and national security, civil liberties, and individual privacy.[3] Specifically, the Order proposes the development of regulatory schemes that will “establish guidance for AI development and use across technologies and industrial sectors,” while taking into account “privacy and civil liberty protections for individuals who may be affected by increased access and use, as well as confidentiality protections for individuals and other data providers.”[4] While the Order lacks details on specific regulatory schemes, state legislation and insight from experts in AI provide a basis for what regulations may come. [5]

For instance, on September 28, 2018, California Governor Jerry Brown signed a cybersecurity bill, SB-327, for Internet of Things (hereinafter “IoT”) devices into law.[6] Although the IoT and AI may seem unrelated, there is a clear intersection between the two. [7] Effective January 1, 2020, SB-327 will be the first cybersecurity law for IoT devices enacted in the United States.[8]  Under SB-327, manufacturers of “connected” devices, who sell their devices in California, must equip such devices with “reasonable security features.”[9] Therefore, it would be no surprise if the government echoed certain aspects of SB-327 in future AI regulations to ensure public safety.

However, where state regulations may seem to be lacking, experts in the field of AI have proposed a handful of regulations that would advance research and development of AI while protecting society.[10]  For instance, one main suggestion is the implementation of sector specific regulations.[11]  AI has the potential to affect a wide range of fields; therefore, a single AI regulation and standard would not be as effective as sector-specific regulations.[12]  For example, the AI in air traffic control should have different regulations than the regulations governing autonomous vehicles on the roads.[13]  The push for sector-specific regulations is likely to become a reality given Congress’ inability to pass a blanket Cybersecurity Act.[14]

Yet, maybe one of the most important suggestion experts have made is the funding of AI programs in universities.[15] After all, the Order sets forth the goal of promoting advancements in AI technology and innovation,[16] and increasing funding for AI programs in universities will facilitate a deeper understanding of AI.[17] Further, increasing the funding for AI technologies in Universities will facilitate the incorporation of AI into fields outside engineering and computer science, such as the social sciences and other relative fields.[18]

Ultimately, the Order will help regulate and promote the booming AI industry. Currently, the value of AI-derived businesses throughout the world was estimated to reach 2.6 trillion U.S. dollars by 2020.[19]  Additionally, AI will “likely lead to millions of lost jobs, especially among less-educated workers, and could exacerbate the economic divide between socioeconomic classes in the United States.”[20]  In light of the Order, the AI industry is now likely to grow at an even faster rate while ensuring the public is well protected.  However, accomplishing the goals of the Order will only be possible by learning from AI experts and from proposed state legislation.

[1] Jason R. Baron, New Executive Order on Maintaining American Leadership in Artificial Intelligence­, Nat. L. R. (Feb. 13, 2019),

[2] See Brett Velicovich, Trump’s Artificial Intelligence Executive Order Will Ensure America Doesn’t Lose the AI Race to China, Fox News (Feb. 17, 2019), (noting the President’s Order is in direct response to China’s plan of becoming the world leader in AI).

[3] Exec. Order No. 13,859, 84 Fed. Ref. 3,967 (Feb. 14, 2019).

[4] See id. (recognizing the Executive Order lacks specificity).

[5] Eliza Strickland, 4 Experts Respond to Trump’s Executive Order on AI, IEEE (Feb. 12, 2019), (“While sparse on details, the order gestures in some potentially positive directions, such as supporting our 2018 recommendation for sector-specific AI regulation. . . .”).

[6] Weak, Default Passwords Banned Under New California Law, MeriTalk (Oct. 8, 2018, 3:44 PM),

[7] See Mark Jaffe, IOT Won’t Work Without Artificial Intelligence, WIRED (last visited Feb. 17, 2018), (finding AI will be used to analyze the vast amount of data produced by IoT devices).

[8] James Mariani, Your Vacuum Cleaner, Your Coffee Maker, and Your Baby Monitor May Be Watching You, So They Better Be Secure: California Passes New Connected Device Cybersecurity Law, Frankfurt Kurnit Klein + Selz (Oct. 8, 2018),

[9] S.B. 327, Chapter 886, Reg. Sess. (Cal. 2018).

[10] Meredith Whittaker et al., AI Now Report 2018, AI Now Institute at 4 (Dec. 6, 2018),

[11] Id.

[12] See id. (explaining a national AI regulation will struggle to meet the sectoral expertise requirements needed for nuanced regulation).

[13] Id.

[14] Adi Robertson, California Just Became the First State with an Internet of Things Cybersecurity Law, The Verge (Sept. 28, 2018, 6:07 PM)

[15] Meredith Whittaker et al., supra note 10, at 5.

[16] Exec. Order No. 13,859, 84 Fed. Ref. 3,967 (Feb. 14, 2019).

[17] Id.

[18] Id.

[19] Charlie Osborne, Artificial Intelligence Will Be Worth $1.2 Trillion to the Enterprise in 2018,ZDNet (Apr. 25, 2018),

[20] Steven Overly, Artificial Intelligence Could Cost Millions of Jobs. The White House Says We Need More of it, The Wash. Post (Dec. 20, 2016)

Share this post