NEWNow you can listen to Fox News articles!
Be afraid. He was very afraid.
That’s the burning message in the media-tech world when it comes to artificial intelligence (AI).
This column, in its own right, is written by a person with no technical assistance.
It’s very rare–once in a blue moon–that I read a piece that completely changes my perspective on a story.
Like many people, I have viewed the rise of AI with a mixture of concern, skepticism and wonder.
DEMOCRATS DISMISS AI FOR MAJOR MESSAGING PROBLEM
It’s fun to combine images in ChatGPT, for example, and I find that some people use it for hyperspeed research. But then you hear anecdotes about AI that involve math problems or spewing nonsense.
Sure, we’ve all seen the warnings that this fast-growing technology will cost some people their jobs, but I thought that would be in Silicon Valley. The era of air travel did not wipe out passenger trains or buses, although they were curtains for the horse and carriage business.
But now comes Matt Shuman, who works in AI, and he’s not just joining the forecasting sweepstakes. He tells us what is happening right now.
Last year, he says, “the new techniques for building these models opened up a faster pace of progress. And then it came even faster. And faster again. Each new model was not only better than the last … it was better by a wide margin, and the time between the release of a new model was shorter. I was using AI more and more, going back and forth with it little by little, using logic to carry my own things.”
On Feb. 5, two major companies, OpenAI and Anthropic, have released new models that Shuman likened to “where you realize that the water has been rising around you and now it’s in your chest.”
Derogatory information has made ChatGPT more accurate. Modest ones scored low. The tone changed the result. (Kurt “CyberGuy” Knutsson)
Bingo: “I am no longer needed for my actual technical work of my work. I explain what I want to be built in plain English, and it just … appears. It is not a difficult framework that I have to prepare. A finished thing. I tell the AI what I want, I walk away from my computer for four hours, and I get the job done. I did well, I did better than I could have done by myself, without the correction of the last few months, with the necessary Aguide. so, I do the planning Now I just I explain the result and I leave.”
Wait, there’s more. The new GPT model “wasn’t just doing my bidding. It was making smart decisions. It had something that felt, for the first time, like judgment. Like taste. That fuzzy sense of knowing what’s the right call that people always say AI can’t have. This model has it, or something close enough that the difference starts to matter.”
This goes beyond the smart world of techies, in case you’re feeling insecure. “Law, finance, medicine, accounting, consulting, writing, design, analysis, customer service. Not in ten years. People who build these programs say one to five years. Others say less. And looking at what I’ve seen in the last few months, I think ‘less’ is more likely.”
AI ENHANCES NATIONAL SECURITY PROBLEMS. HERE’S HOW TO FIND IT RIGHT
My knee-jerk reaction is, okay, I’ll be okay because no super-smart bot can talk about the news on TV or podcasts with the same attitude and verve that I talk about. Then I remember, even as a writer, that news organizations rely heavily on AI.
What about artists who bring soul to rock ‘n roll or bop to their pop? Well, the most famous AI artist is Xania Monet. Some fans were surprised to find out that it wasn’t her, even though she was created by a real poet, Telisha “Nikki” Jones, and most of the audience didn’t care. In fact, “Xania” now has a recording contract worth millions of dollars.
Another sobering thought: “Dario Amodei, perhaps the most security-focused CEO in the AI industry, has publicly predicted that AI will eliminate 50% of white-collar jobs within one to five years.”
Gulp.
Experts predict that AI will eliminate 50% of entry-collar entry-level jobs within one to five years. This figure comes as concerns related to job security are increasing in technology. (Cheng Xin/Getty Images)
This hit the media echo chamber, from Axios to the New York Times to the Wall Street Journal, among others.
The fact that Matt Shuman presents this in a measured tone, not a sky-falling shout, adds to his credibility.
Anthropic, in turn, issued a study that defended its Claude Opus model, “from any attempt to automatically exploit, manipulate, or disrupt” the company’s operations “in a way that raises the risk of future catastrophic consequences.”
The report added: “We do not believe that it has an associated dangerous purpose that would raise the risk of destruction, or that its deceptive power rises to the point of invalidating our evidence.”
95% OF FACULTY SAY AI MAKES STUDENTS RELEVANT TO TECHNOLOGY FOR LEARNING: SURVEY
Meanwhile, the National Review offers a counterweight to what it calls “doomerism.”
First, “many forecasters expect AI to be a bottom-up disruption rather than a bottom-up phenomenon.”
On the other hand, writes Noah Rothman, “there is almost no need to talk about undesirable consequences short of tragedy.
And what about positive impact?
Concerns surrounding AI have led to the rise of “doomerism.” While experts say the “decency and wisdom” in AI talk “doesn’t go viral.” (Stock)
“Instead of eliminating all sectors, it is possible that the workers displaced by AI will be retained in the sectors in which they are already employed.
It’s counterintuitive to think that an industry growing as fast as AI is predicted won’t need human data scientists, research analysts, specialized engineers, and yes, even support and management staff. In addition, sectors such as health care, agriculture and emerging industries will need more human talent than they currently use.
The Conservative magazine is also upset that “participants in this debate do not agree to think that the only solution to the power of dividing AI, regardless of its size, is central government.”
Well, take your pick.
CLICK HERE TO DOWNLOAD THE FOX NEWS PROGRAM
If AI, now able to write well enough to reproduce itself, doesn’t eliminate billions of jobs, or society finds ways to adapt, we can all breathe a sigh of relief.
And if artificial intelligence is as destructive as Shuman’s alarming title already implies, we can’t say we weren’t warned—but maybe we can use it to do our jobs while we work three days a week on three-hour lunches.
I’m not agnostic at this point, except it’s going to be a wild ride.
