Many Americans who have retired in recent years or are approaching retirement grew up listening to The Beatles sing about wanting a revolution that would change the world. Today, such a revolution is taking place.
The launch of OpenAI's ChatGPT in late 2022 brought generative artificial intelligence (AI) into the limelight. Generative AI systems are trained on massive data sets and can generate text, images, audio, and video that can in some cases rival what humans can do.
But while this technology is truly starting a revolution, it's a much different kind than was envisioned in the 1960s. And it could end up changing retirees' worlds in a concerning way.
Image source: Getty Images.
Social Security's main problem
Social Security recipients recently received some really bad news: The program's trust funds will run out of money in 2034. It's no surprise that Social Security is headed for insolvency if major reforms aren't made. However, 2034 is one year earlier than previously forecast.
To understand how generative AI systems such as ChatGPT could impact Social Security, we have to look at Social Security's main underlying problem that's causing it to run out of money. There aren't enough new workers entering the workforce to keep pace with the number of Americans retiring. Social Security is funded primarily through payroll taxes that employees and employers pay.
The Social Security Trustees project that this problem will continue through around 2040. Between 2040 and 2055 or thereabouts, the issue should diminish somewhat as higher birth rates between 1990 and 2008 lead to more new workers. But after then, the Trustees expect the situation to again worsen through around 2078.
There's a key assumption in these projections. Social Security Trustees are counting on the new entrants into the labor pool actually finding jobs. Without enough jobs, payroll taxes won't generate as much revenue as the Trustees are counting on.
Enter generative AI
So how does generative AI fit into this picture? It holds the potential to significantly disrupt the workforce.
Goldman Sachs estimates that up to 300 million full-time jobs across the world could be replaced by generative AI. It projects that as many as two-thirds of all U.S. jobs could be at least partially automated by AI with between 25% and 50% of tasks automated.
An analysis conducted in 2022 by the U.S.-European Union Trade and Technology Council stated that "AI is likely to disrupt large swaths of jobs and tasks." This report even quoted ChatGPT's response to a question about the downsides of AI for workers. The AI app listed several potential downsides, but the first one was: "Job Losses: One of the potential downsides of AI is that it could lead to increased unemployment as machines begin to replace human workers in a variety of industries."
Haven't technological advances in the past created more new jobs than the ones they replaced? Yes. Both Goldman Sachs and the U.S.-E.U. Trade and Technology Council agree that generative AI is likely to create new jobs. The big question, though, is: Will there be enough new jobs to offset the job losses? The best answer at this point is that no one knows for sure.
Different solutions required
It's quite possible that generative AI technology such as ChatGPT poses the biggest threat to Social Security that the program has ever faced. A system built largely on funding from payroll taxes requires a steady influx of new jobs as Americans retire. If those jobs are no longer available, the system falls apart.
There are solutions to every problem. Those solutions will likely be different from some that have already been proposed to bolster Social Security, though.
For example, some politicians have advocated gradually increasing the full retirement age. That might not be as good of an option if many Americans don't have jobs to continue working in until they're older than 67.
One potential alternative would be to tax companies that use AI systems to replace jobs at the same rate they would have paid if they employed people in those jobs. This would at least keep Social Security's funding from being negatively impacted, although it wouldn't address the bigger unemployment issue.
Of course, the current problems for Social Security need to be fixed before the possible problems presented by generative AI are addressed. But hopefully, The Beatles' lyrics in their song "Revolution" will be proven accurate: "Don't you know it's gonna be all right."
