With the technology introduction of the personal assistant Siri and the promise of a completely self-driving car from Tesla soon, artificial intelligence (AI) is radically changing the modern world. AI is an umbrella term to describe advancements that allow machines to learn from experience, adjust to new inputs, and perform human-like tasks. AI learns and improves itself with the large amount of data created through use of the internet, and therefore will continue to advance at an almost uncontrollable pace.
Even with looming AI advancement, limited AI policy exists in the U.S. Because the U.S. government is behind the rest of their world in forming AI policy, AI developers are charged with policing themselves. However, economic incentives pressure these same developers to innovate as fast as possible without “wasting” time on expensive risk analysis. This haste, unchecked by government oversight, may be a detriment to the American people.
While the U.S. government remains disengaged with AI policy, we, as a country, are not prepared for further AI development. We need to get in front of the exponential progression of AI by crafting effective policy that addresses (i) formalizing a structure for increased U.S. government involvement in AI progression, (ii) job security, and (iii) consumer privacy.
Government needs to be more involved in AI development
France, China, and much of the European Union recently recognized the promise and power of AI. Each has published policies and plans for future policies that protect consumer privacy, direct AI exploration via funding, and tackle ethical concerns like job displacement. The United States government, however, has yet to release a comparable AI development plan, leaving AI development unchecked. Government involvement in AI planning and policy is necessary, because the government has the unique ability to direct the development of AI for the public good. Without a structure for government involvement, politicians interacting with AI policy are forced to rely on self-interested private enterprises for information regarding AI news and respective policy recommendations.
One way the government can start shaping AI development is by investing in it, through direct purchases from AI groups that align with the government’s goals or through public grants. Policymakers can allocate funds strategically and encourage projects that benefit the public. Additionally, the government can influence policy through its purchases, exerting considerable pressures for positive AI development from businesses hoping to access the government’s considerable funds.
Government intervention in AI should consider job security
AI is rapidly replacing humans in performing tasks that have historically been performed by blue collar workers, like cashiers, assembly line and warehouse workers. Prior to automation, machines have begun to substitute a fraction of the human workforce since the Industrial Revolution. Even then, industrialization caused a percentage of employees to remain permanently displaced. Experts argue that the consequences of further automation will be more devastating for two reasons: The rapid ability of AI to displace more people permanently all at once (800 million by 2030) and the fact that most job sectors will be affected by AI development and displacement.
The automated workplace could supplant human labor. Therefore, policymakers must work with businesses to foster AI systems that supplement, rather than replace, human skills. Policymakers should ramp up support for programs that provide lifetime retraining and skills development for workers who will transition to what can be a more productive workplace of the future.
Government action should include protecting consumer privacy
Over the last decade, privacy concerns have shifted from those about personal privacy (eg, password theft) to those about institutional control as metadata is now popularly used to garner information about unsuspecting consumers. The shift was realized by many Americans after several data-security breaches, namely the Equifax data breach and the Cambridge Analytica fiasco, which saw people’s data being used in a way they did not consent to. The challenge is not only what information is collected, but also how this personal information is utilized. Firms may easily manipulate potential customers by abusing the naivety of consumers. Bots can gain one’s trust to extract personal information. As many are aware from recent news cycles, politicians and political operatives can target messages, sometimes even misleading ones, in an effort to sway collective public attention.
AI promises improvements to many aspects of society in a variety of contexts, like increasing the efficiency and production of manufactured items, increasing home security, and decreasing energy usage with smart homes, and improving healthcare effectivity with accurate telemedicine. However, for these improvements to come to fruition, government involvement is necessary. Government involvement can ensure AI development benefits and protects all. This November, let’s vote for people who recognize and support favorable AI development.
Madelaine Wendzik currently serves as an Associate Editor for the News and Policy Team at Athens Science Observer and is a Ph.D. student in the Neuroscience Program at the University of Georgia studying neuroinflammation and immune response in pediatric traumatic brain injury. She enjoys board games, downloading one too many podcasts, and anything to do with white chocolate macadamia nut cookies. You can email her at MWendzik@uga.edu or follow her on twitter @SciPolicyGirl. More from Madelaine Wendzik.