Live data from Hacker News

AI is hitting entry-level jobs hardest, Stanford study finds

arstechnica.com

61–70 of 188 posts

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#61
post #32

It's brutal, my son with a CS degree hasn't been able to get to single live interview ina year. A couple of screens has been it. I've reached out to my network, and the reply is always we're not hiring any juniors.

I don’t understand this logic. If I’m being honest, the most productive people with AI on my team are the two youngest. They don’t have product wisdom, yes, but the gap is so noticeable that I pushed my manager to forego the full-time work experience requirement that we have always had so that new hires don’t even know the old way of working and can just reframe the field in terms of AI. AI is the biggest threat in o…

Same here. The absolute star in my team is the youngest. But! I think what's happening is that the distribution in this age group has shifted. Essentially the absolute best (high IQ, super ambitious, great communicator, cracked builder types) are now even better positioned than ever before. But the majority of mass in this age category now suffers, i.e. everyone that's "good" or "very good" but not exceptional/brilliant.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#62

Earlier quoted context omitted.

I don’t understand this logic. If I’m being honest, the most productive people with AI on my team are the two youngest. They don’t have product wisdom, yes, but the gap is so noticeable that I pushed my manager to forego the full-time work experience requirement that we have always had so that new hires don’t even know the old way of working and can just reframe the field in terms of AI. AI is the biggest threat in o…

But the AI transition is different than old technological changes in that you don't really have to learn anything. You don't have to memorize keywords and new concepts like you'd have to for, say, a new JavaScript framework. If you don't know something about AI, you can just ask it and it will tell you. Basically the only hard part for old people is getting USED to using AI. Some haven't, but it only takes a few week…

>If you don't know something about AI, you can just ask it and it will tell you.

If your questions are sharp, probing for detail, and intelligent. Have you seen how people question AIs? It's as if they can't even conceive of the AI being able to supply specific detail.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#63

So we have a crunch of expiring/retiring workers on the senior end and a cutoff of the flow of new talent from the junior end. They’d better pray that we see automation gains above and beyond what we’ve seen so far or else we’ll be paying the price in a single generation.

My really big worry is that the people making these choices just do not care about a single thing beyond this single generation Not a psychologist but my take on most of the wealthy people in the world is they are essentially narcissistic hedonists. They only care about growing wealth as a means to the end of maximizing the amount of fun they can have and to keep their egos happy that they have the most. They don't s…

[dead]

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#64

Partly this is because the big companies are diverting vast amounts of money towards building data centers, instead of hiring new grads. I expect young people who learn AI skills will be fine, once the data center spending binge is over.

What does "learning AI skills" even mean? You're typing into a prompt box. You are not applying any skill, the model is. At best, the skill you need is discernment about _what_ to type in this magical box. But if that gets cheaper and the output of your company doesn't necessarily scale horizontally that well. Why would I need as many people as before?

I imagine the meta will be “build a micro saas product that’s marginally successful” for younger folks. It’s trivial to make a product but you learn a lot seeing what the AI does wrong.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#65
post #5

The article assumes that jobs that can be automated are vulnerable. But in the postmortem of the AI driven layoffs so far, it's more that expectations about AI and productivity are driving the job cuts. In other words shareholders and senior management are being shown the appearance of expectations about AI being met, whether or not it's working. Entry-level jobs are an obvious place to look things that can be cut to…

Mortgaging the future to pay for the present. All of these executives are operating under the pretense that they won’t be held accountable when in 10-15yr their middle bench is a disaster, and they’re probably right.

These executives are going to get some short-term performance based payout and move to another company while talking up the "Bold AI transformation" they lead waaaay before any fallout kicks in. Notice how there's less "We moved to Cloud / on-prem" slop on LinkedIn? It's because the same people have a new bandwagon to chase.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#66
post #3

I wish there were studies on when this situation will normalize. Will the lost jobs ever come back? Or not?

You have market theory saying that at some point, you will not have enough seniors because no juniors were trained. So seniors will become more and more expensive due to lack of supply. At some point, it will be cheaper to take the risk and train the junior entry person again, and things will stabilize out. The sad part is that there will be a generation of 10 years or so of lost juniors while seniors are abundant.

>"At some point, it will be cheaper to take the risk and train the junior entry person again"

I think they will have big problem finding people who will actually train those juniors by then.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#67

i remember seeing the stats presented on moonshots podcast about 3 months ago. they said every quarter, the percentage of CS fresh grads decreased their placement roughly 10% or so and it was down to about 19% at last count and entry level salaries down from 96K to 61k, if i recall.

So revert to more of a mean for white collar work in other words.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#68

Partly this is because the big companies are diverting vast amounts of money towards building data centers, instead of hiring new grads. I expect young people who learn AI skills will be fine, once the data center spending binge is over.

What does "learning AI skills" even mean? You're typing into a prompt box. You are not applying any skill, the model is. At best, the skill you need is discernment about _what_ to type in this magical box. But if that gets cheaper and the output of your company doesn't necessarily scale horizontally that well. Why would I need as many people as before?

Upvoting this for asking the candid question that so many commenters dance around. I can't imagine a 22-year-old confidently walking into an interview (or getting introduced to a team) and assuring everyone that they have the "harness" or "guardrails" to confidently contribute at the level of a 10x engineer.

If we apply the Gartner Hype Cycle to this disruptive technology, I do believe AI will hit a "slope of enlightenment" where an AI-assisted Analyst is properly understood in industries and can take productive roles. Conveniently, this will probably coincide with a lower interest rate where companies have funds for a talent acquisition spree/rapid growth.

With that being said, today's graduates are stuck in an awful market with a lot of charlatans selling crappy online courses and bad ideas.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#69
post #56
post #32

It's brutal, my son with a CS degree hasn't been able to get to single live interview ina year. A couple of screens has been it. I've reached out to my network, and the reply is always we're not hiring any juniors.

This sounds much like the civil engineering crisis in my region of the world after the 2008 financial crisis. There was barely any new construction going on for a couple of years and once it started anew, there was a gap as while people were retiring, new hires weren't replacing them. I guess we're just going to shrink as a field until the market won't be able to handle it.

Our field started as a niche for geeks. For a long time, CS paid you the average for engineering work - hardly equivalent to six figures today for most with the exception of seniors. It's still like that on most of the world.

Many people here lived through ridiculous times. It was possible to make 1M in a year without incredible skills/hard work. Big tech was and still is full of these people.

I'm not saying that was a bad thing at all, but it always felt unsustainable. CS salaries will decline and fewer people will want to be software developers, balancing the market. Then it may go back to where it came from with the geeks sticking around. No more coding bros. Maybe that's okay.

When I got my degree, I never expected more than a slightly above average salary. I do this because I love it and can't imagine doing anything else. I was lucky and I'm thankful for the luck I had.

Re: AI is hitting entry-level jobs hardest, Stanford study finds

#70
post #5

The article assumes that jobs that can be automated are vulnerable. But in the postmortem of the AI driven layoffs so far, it's more that expectations about AI and productivity are driving the job cuts. In other words shareholders and senior management are being shown the appearance of expectations about AI being met, whether or not it's working. Entry-level jobs are an obvious place to look things that can be cut to…

Mortgaging the future to pay for the present. All of these executives are operating under the pretense that they won’t be held accountable when in 10-15yr their middle bench is a disaster, and they’re probably right.

>All of these executives are operating under the pretense that they won’t be held accountable when in 10-15yr their middle bench is a disaster, and they’re probably right.

If everyone's middle bench is a disaster, there's no one that did it right to contrast their poor decision-making. They're hiding the evidence that will help them skip accountability.

Post reply on HN