Can U.S. Civic Culture Survive the AI-Internet Age? – Inside Sources

There was a time, not that long ago, when curious voters could know where political candidates stood on important issues of the day by reading their local newspapers.

Those days are gone, as are many of the newspapers. AI Overviews are no substitute.

According to a recent report from Northwestern University’s Medill School of Journalism, “almost 40 percent of all local U.S. newspapers have vanished” during the past two decades. This has resulted in what the authors of a recent book, “News Hole,” describe as a “massive reduction in … reporting capacity” nationwide.

Here’s one example: In 2005, the Providence Journal, Rhode Island’s largest newspaper, had a staff of 150 to 160 editors and reporters. By December 2022, the Boston Globe reported, the Journal’s news staff was down to 15 to 20.

From a civic standpoint, the decline in reliable news reporting has been devastating, leading Cambridge University Press, publisher of “News Hole,” to lament that “residents of cities and counties across the United States now have access to less coverage of their local governments than at any time in modern American history.”

Many optimists think the internet, which completely changed the way people consume news, would become a superior replacement to newspapers. After all, information about candidates — their political party, education, family and policy positions — along with criticisms from their opponents, would be available to us 24/7 through a simple Google search.

However, the internet has proven to be a poor substitute for newspapers. Instead of clarity, it provides noise — and now, in its latest iteration, AI slop.

What hasn’t changed is the need for citizens to make informed decisions about candidates and issues.

Immersed in the Bible and the writings of the world’s greatest thinkers — Homer, Aristotle and Plato, among others — America’s Founders knew that the free society they envisioned, a democratic republic, could endure only if citizens were educated in the details and parameters of American self-government, and in the virtues of good citizenship.

The Founders understood that to exercise the responsibilities of citizenship properly — not only with respect to voting but also with respect to the rights of others — citizens would need the ability to judge options soberly, including practical and ethical perspectives.

As our forebears understood, citizenship is more than showing up on Election Day. The First Amendment exists for a reason: so we can learn, brood, debate and understand.

The internet and AI serve many useful purposes. Promoting good citizenship isn’t among them.

The internet is a free-for-all, not a platform for information alone but for disinformation, misinformation, propaganda, hooking up, photo-sharing, you name it. If nobody is there to report on what’s going on in your city council, or in the deep bowels of a federal agency, you won’t find any information online. And AI, of course, is only as good as the information it is trained on.

In addition, neither has a conscience nor a moral compass, though some internet platforms purport to screen out the most egregious bad actors (and, as we know, do a poor job of it, often screening out information and opinions the screeners disagree with).

If prompted, AI can summarize Homer’s great epic poems, “The Iliad” and “The Odyssey,” but it can’t teach you how to distinguish right from wrong, how to read the facial reaction of the person sitting across from you, or how to judge the value — to you! — of a policy or a purchase your city government is considering.

Worse, AI’s large language models are obsequious and flattering, and if you ask them to provide support for your point of view on a given matter, they will do just that, even if your view is harmful or incorrect. That’s no way to make decisions in a free society.

Citizenship requires much more than writing the correct prompt for a chatbot. Developing the judgment and discernment the Founders envisioned requires study of economics, ethics, history and literature.

This provides a baseline. Good choices require even more: reliable and nuanced information.

AI isn’t there yet.