In recent years, artificial intelligence (AI) has created significant advances from facial recognition to medical diagnosis. Yet, despite significant achievements in AI, the former Pentagon software chief, Nicolas Chaillan, warned the U.S. is falling behind other countries in terms of AI development, and “the failure to respond [is] putting [Americans] at risk.” While Chaillan is talking specifically about military uses for AI, his warning is a reminder that the government should recognize the vast array of benefits that AI can bring for consumer welfare and promote a policy environment that supports innovative technologies.
The European Union (EU) serves as a great example of how excessive regulation has resulted in technological stagnation. Recently, the EU passed the Artificial Intelligence Act of 2021, marking the beginning of EU-wide unified rules on AI. The act outlines regulations on medical devices and facial recognition. Under these new regulations, the EU requires high-risk AI tools to “comply with certain mandatory requirements” before they are allowed on the market, and fines for violating the law can reach 6% of a company’s total revenues. Regulations like these create a hostile environment especially for startups in the AI space.
In the US, Maine seems to be following the EU’s lead by introducing the most stringent laws regulating government AI to facial recognition technology. L.D. 1585 bans most uses of facial recognition technology by public officials such as law enforcement. By outlawing facial recognition technology, however, the security and safety of Maine residents are threatened, making them more susceptible to criminal activity.
Currently, no US federal regulations govern how private companies or governments use AI. The absence of any statute has raised concerns about the potential misuse of AI, with democrats concerned that its use could amplify racial discrimination and conservatives worried about infringing constitutional protections. These concerns open up the possibility that Congress and states would impose overly stringent regulations that deny consumers the benefits of AI. One particular concern is that these onerous regulations could disincentivize AI start-ups from bringing new and innovative technologies to the market.
AI has also sparked fears that it will imperil millions of jobs. Despite popular assumptions, empirical studies demonstrate that the use of AI is an important driver for economic growth, increasing income for Americans. It is estimated that AI will create over 58 million jobs and add $1.2 trillion to US GDP in the next five years. Without an open regulatory environment that nurtures AI innovation, Americans will be denied access to advanced technologies, lagging behind nations like China.
Among the multiple applications, AI technology has aided law enforcement in criminal investigations, resulting in enhanced consumer safety. For example, in 2015, Thorn, a nonprofit dedicated to protecting children from sexual exploitation, built Spotlight, an AI tool that uses facial recognition to help investigators find victims of online sex trafficking. There have been over 1,980 cases of sex trafficking identified by law enforcement officials since Spotlight was launched. Additionally, by using this technology, law enforcement agencies have been able to conduct 7,422 investigations. Over-regulating AI and prohibiting the use of facial recognition technology would limit the ability of law enforcement to track down criminals.
AI has also been widely deployed in healthcare to allow doctors to diagnose patients faster, improving health outcomes. For example, acute kidney injury (AKI) can be life-threatening if not treated and accounts for 11% of hospital deaths. Early detection is crucial in preventing AKI from progressing. The Department of Veterans Affairs and DeepMind Health collaborated in 2019 to develop a Machine Learning (ML) solution that can predict AKI up to 48 hours before patients display symptoms. As a result, more than 90% of AKI cases are currently identified and treated faster than by conventional procedures, leading to improved health outcomes.
While the temptation to impose stringent regulations on AI is great, lawmakers should be cautious. Imposing stringent regulations on the development and use of AI will not only see the US fall further behind other countries, but it will also leave Americans more vulnerable to crime and facing poorer health outcomes.
The only way this can be avoided is if the federal government crafts a policy environment that nurtures innovation, not snuffing it out.