The AI Apocalypse

I have to write this here, as I don’t want to get fired … again.

Alas, it is crystal clear that the recent graduates in fields that have been in high demand is dropping.

For more than a decade, the wisdom was to go to as good of a school as you could get into, and study CompSci and add tons of mathematics and statistics to prepare for a gravy train in the Tech ecosystem.

Alas, that seems to be at the end, as this article in The Atlantic posits: “The Computer-Science Bubble Is Bursting” (gift link)

The job of the future might already be past its prime. For years, young people seeking a lucrative career were urged to go all in on computer science. From 2005 to 2023, the number of comp-sci majors in the United States quadrupled.

That is a whole lotta students, many of whom jumped into the Machine Learning field that evolved into the current crop of AI (Generative AI) researchers, developers, and product sages.

I mean, how sucky is is that you work your heinie off to get into a top tier uni to study for one of the holy grail jobs, Programmer for one of the tech giants coding algorithms to drive more engagement and addiction amongst social media users, or now, in the wunderkind areas that Antrhopic, OpenAI and others are fielding.

At Stanford, widely considered one of the country’s top programs, the number of comp-sci majors has stalled after years of blistering growth. Szymon Rusinkiewicz, the chair of Princeton’s computer-science department, told me that, if current trends hold, the cohort of graduating comp-sci majors at Princeton is set to be 25 percent smaller in two years than it is today. The number of Duke students enrolled in introductory computer-science courses has dropped about 20 percent over the past year.

In a way, I have to laugh because starting in the mid to late ‘aughts, the collective wisdom was that you couldn’t go wrong studying Computer Science, and pursuing your Ph.D. in that field.

Fresh graduates were pursued by the tech majors (particularly the Magnificent 7), and the Covid era seems to make them even more bulletproof fields.

Especially since the formula for success was to get accepted to a top tier Comp Sci program at an elite university, and attend just long enough to find a buddy to found a company with, and to then drop out, start the next big thing, grow at all costs, and then cash out and become VC royalty (or you desire to be like Zuckerberg, basically unassailable at the top of a massive company).

Meet Chriss Gropp:

… a doctoral student at the University of Tennessee at Knoxville, has spent eight months searching for a job. He triple-majored in computer science, math, and computational science at the Rose-Hulman Institute of Technology and has completed the coursework for a computer-science Ph.D. He would prefer to work instead of finishing his degree, but he has found it almost impossible to secure a job. He knows of only two people who recently pulled it off. One sent personalized cover letters for 40 different roles and set up meetings with people at the companies. The other submitted 600 applications. “We’re in an AI revolution, and I am a specialist in the kind of AI that we’re doing the revolution with, and I can’t find anything,” Gropp told me. “I found myself a month or two ago considering, Do I just take a break from this thing that I’ve been training for for most of my life and go be an apprentice electrician?”

Yikes, imaging all that work and contemplating a journeyman program to learn to be an electrician.

But why is it suddenly a struggle for these young, very educated and quite talented (you don’t get into the Comp Sci programs at Stanford, Duke, Princeton, or Carnegie-Melon without a lot going for you) to find employment?

In the ultimate irony, candidates like Gropp might be unable to get jobs working on AI because AI itself is taking the jobs. “We know AI is affecting jobs,” Rusinkiewicz, from Princeton, told me. “It’s making people more efficient at some or many aspects of their jobs, and therefore, perhaps companies feel they can get away with doing a bit less hiring.”

Ah yes, AI.

AI is coming for coding jobs, in a big way. My own company is expecting a full 70% of our code to be written by AI in the next two years. I personally think that is hazardous, but it seems to be very common in this realm:

Tech leaders have said publicly that they no longer need as many entry-level coders. Executives at Alphabet and Microsoft have said that AI writes or assists with writing upwards of 25 percent of their code. (Microsoft recently laid off 6,000 workers.) Anthropic’s chief product officer recently told The New York Times that senior engineers are giving work to the company’s chatbot instead of a low-level human employee. The company’s CEO has warned that AI could replace half of all entry-level workers in the next five years. Kinder, the Brookings fellow, said she worries that companies soon will simply eliminate the entire bottom rung of the career ladder. The plight of the tech grads, she told me, could be a warning for all entry-level white-collar workers.

My World – Training

I work in the realm of technical training, in specific IT training.

Here, we don’t do much coding, but we do build a lot of content to prepare people for certifications. We have entry level to expert level, and all steps in between.

My fear is that AI is going to hollow out the mid career people.

Someone with an “associate” level set of knowledge (because you need to know some foundational skills) plus a custom RAG’d Chatbot and you no longer need the “professional” levels.

This development means that in ten years (or less) there will be almost nobody on that career ladder. You will have tons of entry level people augmented with AI tools, and a small corpus of experts (those you call when the shit really hits the fan) and nobody in the middle.

As the experts “grey out” (a.k.a. retire) there will be a dearth of talent to replace them, because, a skilled person plus AI is much less than a true expert.

It’s worse in marketing

In my world it is bad. In general marketing, the people who write copy, and build out campaigns and the like, all that work is being subsumed by AI already. The number of entry level jobs in that space is already about nil.

And freelancing is getting hammered, as corporate marketing drones are satisfied with the GenAI slop that the current crop of bots crank out.

That is a field that is dooooooomed.

Last thoughts

A final pull from the article:

Whether the past few years augur a temporary lull or an abrupt reordering of working life, economists suggest the same response for college students: Major in a subject that offers enduring, transferable skills. Believe it or not, that could be the liberal arts. Deming’s research shows that male history and social-science majors end up out-earning their engineering and comp-sci counterparts in the long term, as they develop the soft skills that employers consistently seek out. “It’s actually quite risky to go to school to learn a trade or a particular skill, because you don’t know what the future holds,” Deming told me. “You need to try to think about acquiring a skill set that’s going to be future-proof and last you for 45 years of working life.”

I guess we should see a revival in the humanities and social sciences. But somehow I am doubtful.

I am glad that I am a few years from retiring, and will not have to ride this roller coaster.