LENOX — I was planning to let ChatGPT write my Bottom Line column to see if anyone noticed. Actually, from what I’ve been learning, the artificial-intelligence “author” likely would turn out a fully professional article — though lacking in style, whimsy and assemblage of facts beyond the Wikipedia level.
The following is my own. But it’s on the subject of AI which, like it or not, is invading our lives more rapidly than might have been expected.
People in the word trades — authors, journalists, public relations writers — are in a frenzy, understandably. Much of the hand-wringing is justified. An author and newspaper columnist told me he was shocked to find that ChatGPT could produce certain material equal to, or even superior, to his own.
After reading dozens of news articles on this topic, I remained flummoxed.
Seeking some understandable analysis, I encountered an NPR interview with Jack Clark, co-founder of the AI company Anthropic. According to its website, the AI safety and research startup — founded by former members of OpenAI — is working to build reliable, interpretable and steerable artificial intelligence systems.
“My No. 1 concern about AI right now is AI systems can do more things than their creators know that they can do,” Clark told “Morning Edition” co-host A Martinez. “It’s kind of like if we were in the business of making cars. After you release the car, someone discovers it can fly or go underwater, and you had no idea as the car manufacturer.
“That’s where AI is today. Systems get released. Then some 17-year-old with a laptop discovers that the system can do a completely wild thing that its creators did not anticipate.”
Speaking of wild things, financial markets were shaken up briefly on Monday when Facebook, Twitter and the Kremlin-controlled RT website circulated an image of black smoke supposedly erupting from a government building near the Pentagon.
It was an AI-generated fake that played into concerns about how artificial intelligence could be misused for evil results, producing disinformation resulting in panic, such as a stock market crash.
The image was deleted quickly, but, as The New York Times reported, “the incident underscores how even unsophisticated spoofs can spread misinformation quickly, especially via trusted social-media channels.”
Government agencies are issuing warnings: Securities and Exchange Commission Chairman Gary Gensler has pointed out that bad actors could use AI to exploit “the fragility of financial systems.”
And the Federal Trade Commission is alarmed about how AI-generated deep-faked images and cloned voice systems could be used to trick people in new kinds of fraud schemes.
Technology guru Bill Gates, co-founder of Microsoft, told a conference in San Francisco recently that an AI personal assistant would have a profound impact. “You will never go to a search site again,” he suggested. “You will never go to a productivity site. You’ll never go to Amazon again.”
Looking to a “safer AI” future, Clark, the Anthropic co-founder, offered some insight: “Something which most technologists say privately when you talk about AI policy, in two or three years the systems are going to be far more powerful, and we can’t really anticipate them today.”
So, what’s his bottom line for avoiding chaos?
“When people are regulating this technology, they’re treating it like a normal technology which evolves relatively slowly and relatively predictably,” Clark pointed out. “This technology evolves very quickly and relatively unpredictably. So if anything, my main takeaway is the future is going to be a lot weirder than the present, and we should have our minds kind of pointed towards that, as well as dealing with these challenges we have today.”
I don’t want to be too alarmist, so I was curious to see whether ChatGPT could pull our nation from the brink of financial disaster.
After opening an account, I asked my new AI agent to offer a solution that could resolve the crisis through a compromise acceptable to Congress and the White House.
The results? Increase the debt ceiling. Prioritize spending. Cut spending. Negotiate a bipartisan agreement.
These options, obvious to anyone following the torturous hour by hour news coverage, were described in summaries that were clear and concise, pointing out that “this often requires bipartisan cooperation and consensus-building.”
And ChatGPT reminded me: “It’s important to note that these actions can vary depending on the specific circumstances.”
Impressed by this effort via the free app, I should consider the more powerful version available for $20 a month.
The intelligence may be artificial, but it surpasses the actual level of thinking we see on display from House Speaker Kevin McCarthy’s narrow right-wing majority.
I’ll have to ask ChatGPT who it supports in the 2024 election.