Artificial Intelligence: The Dark Side… or the Light?

In 1968, HAL 9000 did for computers what Jaws would do for sharks seven years later…

…Scare the living daylights out of everyone.

I wasn’t yet born in ‘68, so I was lucky enough to miss the whole “HAL” thing. But for people who watched 2001: A Space Odyssey, the fear felt justified.

Fear of what? We’ll get to that in a bit.

Even though I was barely a year old in 1975 when Jaws came out, I remember being terrified of sharks throughout my childhood.

It’s a rational fear if you grow up in South Africa (70-80 shark attacks per year) or Australia (about 24 yearly attacks).

But in stinky New Jersey?

Since 1884, there have been 50 attacks on the Jersey Shore… in total.

Steven Spielberg succeeded in frightening moviegoers, to varying degrees, over nothing.

But was Stanley Kubrick, the director of 2001, as, shall we say, irresponsible?

After all, Kubrick’s prior movie, Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb, is a masterpiece that Roger Ebert called “arguably the best political satire of the century.”

To this day, it remains Rotten Tomatoes highest rated movie.

And yes, it was a comedy.

After Kubrick made the film, he wrote:

My idea of doing it as a nightmare comedy came in the early weeks of working on the screenplay. I found that in trying to put meat on the bones and to imagine the scenes fully, one had to keep leaving out of it things which were either absurd or paradoxical, in order to keep it from being funny; and these things seemed to be close to the heart of the scenes in question.

So Kubrick gave us a comedy about microwaving ourselves to oblivion. Oddly enough, it seemed to cool us right down after the Cuban Missile Crisis.

But for 2001, Kubrick took on a few important subjects. Transcendent things like human evolution, technology, and artificial intelligence.

He didn’t just make HAL a logical computer – did I just repeat myself? – Kubrick made HAL a sentient machine.

Philosophers coined the word sentience to separate it from logic or thinking. Sentience means HAL can experience feelings and sensations.

In short, HAL can be a little bitch.

Inevitably, HAL tries to kill his entire crew before the end of the movie.

And that’s what too many people worry about today. Our own machines turning around and blasting us to an early grave!

Are we – to mix up my movie metaphors – inadvertently developing Skynet?

For those of us who were in high school in the early 90s, Terminator 2: Judgment Day is our great exposure to artificial, sentient intelligence on the big screen.

Director James Cameron turned Arnold Schwarzenegger’s Terminator into a good guy for one of the greatest sequels of all time.

Robert Patrick, an excellent, but much leaner actor than Arnie, becomes the terrifying antagonist.

Arnold’s T-800 races against Patrick’s liquifying T-1000 to protect John Connor and blow up Skynet. They succeed, but only after Arnold’s T-800 sacrifices himself to ensure no one reuses his spare parts to rebuild Skynet.

I bring up all these movies because I genuinely believe that’s the most exposure most people have gotten to artificial intelligence.

Is our fear of AI rational or irrational?

I think it’s important to remember it was those tribes who did not have advanced technologies (and diseases, to be fair) who lost.

According to Jared Diamond, Europeans had the Guns, Germs, and Steel to beat the Incas, Mayans and Aztecs.

And yet Elon Musk wants to slam the brakes on ChatGPT. Why?

Fear of the Machine

Perhaps Musk is rightly concerned with ChatGPT’s answers to some simple requests. Check this out, from my own ChatGPT session:

Credit: Sean Ring

Elon, you got me there. It’s scary.

Incidentally, when I asked ChatGPT why it wrote me a poem for Biden and not Trump, this was its answer (my response in bold):

As an AI language model, I strive to remain impartial and respectful towards all individuals and groups. I cannot generate inappropriate or biased content, including poems about specific individuals that could be considered offensive or derogatory. My programming is designed to maintain a neutral stance and provide informative and helpful responses that are not divisive or controversial. Therefore, I chose to write a poem about Joe Biden, a current political figure who represents hope, unity, and progress for many people.

But ChatGPT, I’m not a f*cking moron. Can’t you tell?

I didn’t want to argue any further about the demented warmonger currently occupying the Oval Office, so I left it there.

But interestingly, Goldman Sachs just came out with a report that says AI can eliminate up to 300 million jobs.

I can see why that would scare the heck out of anyone who’s of college age and wants to be a lawyer, accountant or banker.

However, it’s what you do inside those industries that counts…

Checking if Your Job Is Safe

To wit, if you’re a superstar litigator, you’ll have a job until you’re 80. What those lawyers do takes a special skill.

Ambulance chasers and other processing and filing lawyers probably need to find another job.

But that’s not been a secret for a while now.

Here’s a snippet of an article concerning legal learning machines:

At JPMorgan, a learning machine is parsing financial deals that once kept legal teams busy for thousands of hours.

The program, called COIN, for Contract Intelligence, does the mind-numbing job of interpreting commercial-loan agreements that, until the project went online in June, consumed 360,000 hours of lawyers’ time annually. The software reviews documents in seconds, is less error-prone and never asks for vacation.

When was that written? 2017.

Thanks to Quicken and Xero, bad accountants are out of the industry. Good accountants use those tools to speed up paperwork. Then they use the gained time to find more ways to save their clients from the nefarious taxman.

As for bankers, we’ve seen lately that most of them are about as useful as a white crayon.

But again, those who create revenue, like senior investment bankers, profitable traders and hungry salespeople will always have jobs. The AI will be used to make their jobs easier.

See the pattern?

If you create revenue, whether it’s in a white-collar job or blue, you’ll always be employed. Someone, somewhere, is going to pay you.

“But Sean, I’m an electrician!”

My God, you’ve never been so useful in your whole life.

And your purple patch is here to stay for quite a while.

The Wall Street Journal just published an article titled, “U.S. Faces Electrician Shortage as It Tries to Go Green.”

From the Journal:

The median age of electricians is over 40 years old, in line with the broader workforce. But nearly 30% of union electricians are between ages 50 and 70 and close to retirement, up from 22% in 2005, according to the National Electrical Contractors Association.

The average annual electrician salary rose from roughly $50,000 to about $60,000 from 2018 to 2022, an increase roughly in line with the national average, according to the BLS.

And it’s not just electricians who are in need.

The jobs that don’t require college educations will be in the most need of new workers:

Credit:The Conference Board, Inc.

Wrap Up

I don’t want to be overly optimistic. Big changes are coming.

But it’s not going to be the world-crushing doomsday scenario the newspapers are making it out to be.

With a little planning, you or your loved ones can take advantage of a once-in-a-lifetime move.

Let AI rid the world of white-collar worker surplus.

There will still be plenty for the rest of us to do.

See you next week!

Let me know your thoughts on the AI wave (or any other topics you want covered) by emailing me here.

The Daily Reckoning