The race to regulate AI: The next frontier for law and society

Artificial intelligence (AI) is set to be the next major technological advancement to dramatically impact modern society. From transforming the way we work, to increasing efficiency in outdated systems, the changes promised by AI have the potential to be utterly transformational. While this brings a huge range of opportunities, there are also some enormous challenges to overcome if humanity is to strike an effective balance between progress and risk.

History shows that society and the law do not always handle rapid innovation well. Take technologies such as the steam engine and automated loom, for instance, where progress was met with varying degrees of resistance and fear before the benefits were fully realized. In the case of AI, harnessing its potential while safeguarding against misuse means legislators must take a measured, risk-based approach to regulation that embraces change alongside effective safeguards.

Lessons from the past

The pattern of resisting and then embracing innovative technology has repeated itself throughout history. Among the most famous historical examples was the introduction of the power loom during the Industrial Revolution, with workers becoming increasingly worried about machines replacing human jobs and labor. In 1826, for instance, three days of rioting in Lancashire saw 21 mills attacked and six protestors shot and killed by the military after refusing to disperse.

Yet, once integrated, these and other technologies enabled enormous progress for manufacturing processes and played a role in delivering huge economic growth. Today, many people believe that AI and machine learning will follow a similar trajectory - as the technologies elevate efficiency and productivity to levels previously unseen in the modern workplace. At the other end of the debate, however, some argue that these quote-unquote ‘sentient robots’ present an existential risk to humanity, raising concerns about where the development of AI will take us in the long term.

Stepping back for a moment, however, it is important to recognize the difference between true AI, which seeks to mimic human-level intelligence, and today's machine learning algorithms that excel at specific tasks. Technologies like autonomous vehicles and chatbots, for instance, use machine learning, not true AI. Despite the hype, it’s likely that technology will not attain human-level intelligence in the near future, and in the meantime, machine learning will quickly bring transformative change.

Balancing risks and rewards

Like any powerful technology, AI carries risks as well as benefits. For example, applications using AI to create personalized education for people with learning disabilities, improve medical diagnoses, and increase accessibility for the visually impaired are likely to deliver enormous, transformative progress. The problem is that more concerning scenarios are also emerging, such as the use of AI-generated phishing emails, hacking tools and deep fakes. As capabilities improve, the threats posed by those with malicious intent will inevitably rise. However, AI is also already being used to counter these risk

As a result, fears about AI unleashing unintended consequences are understandable but must be balanced with a mindset that enables progress. Rolling out hastily drafted legislation, for instance, may require ongoing amendment, during which time AI innovation may be stifled, or the competitive advantage lost.

Ideally, legislators should proceed cautiously in order to determine acceptable levels of risk rather than banning broad areas of research, development or real-world implementation. AI experts and ethicists should also have a seat at the table to inform nuanced policies and regulations. It’s not likely to be an easy process, especially given the definitions of digital privacy, cybercrime, liability and other issues may well require a serious rethink in light of systems that can autonomously generate content and make decisions.

Perfecting legislation for emerging technologies has often been a difficult path, with initial attempts to apply laws to the digital world often falling short, not least because of the sheer pace of change. New challenges posed by AI's almost limitless possibilities and autonomous capabilities will further test existing legal frameworks, underlining the need for technical and ethical experts to advise the policymaking process if future laws are to be appropriate for an AI-powered future.

Ultimately, AI has immense potential to transform our lives for the better, but it also carries risks. As the technology proliferates, stakeholders must stay grounded in facts, and while thoughtful regulation focused on preventing harm is needed, we cannot allow fear to outweigh hope or limit progress. By learning from the past and proceeding deliberately but open-mindedly, we can ensure AI is governed properly and used for good. The widespread integration of this technology presents new frontiers for the law, but with insight and planning, we can have confidence in a future powered by AI.

Photo Credit: Alexander Limbach/Shutterstock

Richard Starnes is Cybersecurity Strategy Director, Six Degrees.