While artificial intelligence is grabbing headlines and enabling new business innovations, its potential negative effects on society are often overlooked. To ensure a better society for all, how can humans regain control and be more involved in the development of AI?

That was the question posed by new research from the University of Michigan, which reviewed the research literature on technologies sitting under the umbrella of the “Fourth Industrial Revolution”, and found that there was a preponderance of tech-enabled business benefits and very little attention given to any societal implications.

“We’re talking about AI the wrong way—focusing on technology, not people—moving us away from the things we want, such as better medications, elder care and safety regulations, and toward the things we don’t, like harmful deepfakes, job losses and biased decision making,” the researchers explain.

“Our new framework is a theory-based attempt to go in a new direction by recentering humans in actions and outcomes in the discourse about ever-smarter machines.”

Societal impact

Using ChatGPT as an example, the researchers point out that while developers may not have intended for their AI chatbot to facilitate cheating or generate recommendation letters, they failed to fully consider the potential social implications of their creation.

Instead, their focus was on the benefits of reduced operational costs. The aim of their study is to help society, including regulators and organizations considering adopting AI, better understand the societal implications of increasingly intelligent machines.

The study highlights the need for a shift in policy-making towards a more proactive approach. Rather than reacting to technologies as they enter the market, laws need to take a broader view of AI and address four key emerging machine capabilities:

• Decision-making: What should be the boundaries of machine decision-making, how can we adjust those boundaries as machines advance, and how can we manage biases and other sources of inaccurate decisions proactively?
• Creativity automation: How can we address job displacement in creative fields like visual illustration and music, who owns the rights to AI-generated works, and how can we manage the risks of perfect and cheap deep fakes of anyone?
• Machine-human relationships: As humans increasingly interact with machines with human-like capabilities, how will workers respond to AI bosses, how might “AI friends” influence democracy, and how might human relationships evolve over time as a result?
• Machine-to-machine collaboration: How can we manage the exponential capabilities of inter-machine collaboration, establish rules for machine interactions that balance positive outcomes with risks, and determine the appropriate level of autonomy while considering the risks of complex inter-machine systems?

Human-centered approach

According to the study, the research community can contribute to this effort by portraying AI as machine capabilities that mimic human cognition and communication. This human-centered approach moves away from the complexity of technology and instead emphasizes what these machines actually do in human terms, and the impact on humans, society, and business.

However, the study notes that much more research is needed to expand on these concepts, including the application of ethical frameworks to support policies for machine capabilities.

The researchers predict that everyone will be increasingly impacted by intelligent machines that emulate human capabilities, whether it’s at home, in the workplace, or in other spheres of society such as the legal system. There’s no reason these machines should be allowed to operate in the world before potential problems are identified. If issues are caught early, they can be corrected or even abandoned, or the machines can be designed from the outset to minimize the likelihood of problems occurring.

“In the short run, our approach may simply limit their negative impacts, but in the long run, it may lead to the development and deployment of AI systems based on their benefits and costs for our society,” the researchers conclude.