AI Should Serve Humans

           I’m a senior citizen still struggling to come to grips with the last generation of technology. I would just as soon not have to think about Artificial Intelligence (AI). It’s here, though, and we’d better get ready. Several resolutions that were approved at our county convention last month address the need for putting guardrails in place.

            One issue that has indeed been getting a great deal of attention is the explosive growth of data centers to power AI. Data centers are massive consumers of electricity and water, and a resolution calling for a Data Center Moratorium notes that increased demand for electricity complicates our ability to transition away from fossil fuels. Communities where they have been proposed are rightly worried about the impact they will have, but local opposition is challenged by corporate demands that local governments sign Non-Disclosure Agreements that seek to limit transparency. We would be foolish to trust corporate promises, and the resolution calls for Minnesota’s legislature to enact strong measures to ensure a thorough review of these projects and their environmental impact. At the very least, the state should not be subsidizing data centers with generous sales tax exemptions.

            Concern has also been growing about the threat that AI poses to the labor force. Apart from the danger that AI could replace workers and cause a major uptick in unemployment, there are also a number of ways in which AI could alter management practices to the detriment of those who still have a job. Our friends from organized labor put through a resolution calling for Worker Protection from AI. To the extent that AI is adopted in hiring, evaluation, and discipline of workers, the resolution recognizes that the whole process can become shrouded from human oversight. Here, too, transparency is key, and the resolution calls for measures to ensure that humans control the machines and not vice versa. In particular, collective bargaining rights must be protected to guarantee that workers are not subjected to arbitrary or discriminatory actions and that employers are held accountable.

            Workers are not the only ones who can be negatively affected by decisions made by AI. Another resolution calls for Ensuring Human Review in Care. The concern is that health insurance companies will adopt AI to review claims using unproven technology. While AI can process massive amounts of data with incredible speed, its decisions will only be as good as the data it absorbs. Will it always discern the difference between medical science and the blathering of RFK Jr.? Patients should have a right of repeal when reasonable claims are denied, which happens often enough with or without AI, but introducing AI threatens to make the process even more impersonal and insensitive. The practice of medicine should remain a matter of humans treating other humans, trusted to medical professionals who understand the art as well as the science of working with people.

            There is a doomsday scenario that AI will make computers smarter than humans and conclude that they should run the world. In the short run, there are more immediate dangers that we are losing control of our lives to machines that indeed only think like machines. Common sense and empathy are in short enough supply among humans, but they remain a human quality that machines have yet to replicate.

Paul Harris