Millions of jobs affected!
A white-collar bloodbath!
Massive economic disruption!
A technological spreading of Marxist propaganda!
Politicians, pundits, and C-suiters everywhere are shouting out about the epic mud stomping that artificial intelligence (AI) is going to do on our stable and beloved job market in the next few years. Or the next ten minutes, depending on whom you’re listening to. They’re loud. They’re apocalyptic in some cases, dystopian in others. And, yeah, they’re freaking people out.
The bad news is coming from all sides, though the numbers vary widely. On the high side, Goldman Sachs predicts that AI could impact up to 300 million full-time jobs worldwide, particularly in the U.S. and Europe.
On the low end, there’s the World Economic Forum, which predicts that by 2027, 83 million jobs will be lost to AI, while AI will create 69 million jobs. That would result in a net 14 million job loss.
Then there’s the guys pitching the real grabbers. Like Dario Amodei, CEO of Anthropic, who says he thinks AI could wipe out half of all entry-level white-collar roles within five years, a situation he’s calling a potential “white-collar bloodbath.”
Or Adam Dorr, a real doom-and-gloomer, who predicts AI will wipe out all (yes, all) human jobs by 2045 except for politicians, sex workers, and ethicists. I’m going to take a wild guess and say that Dorr thinks of himself as an ethicist. And he probably also hasn’t been to CES in the last few years, or he’d have seen what the Japanese have been doing with sex robots. (Hint: disturbing.)
By comparison, our intrepid political leaders’ views on AI seem tame. Even President Trump is sticking to a relatively middle-of-the-road prediction (though he is convinced that AI might be spreading Marxist propaganda). He thinks AI will take between 10 and 15 million jobs in the next five years, but doesn’t believe the government’s role is to restrict AI; instead, it should give tech companies more freedom to use it. (Some of his other beliefs on AI’s long-term effects are more disturbing than the Japanese sex robots, but let’s stick to the topic at hand.)
Let’s all just take a breath, shall we?
What should the rest of us take from all these, let’s be fair, often vague predictions? Nobody knows, that’s what. If AI is going to supplant 14 million jobs, as the WEF predicts, that’s certainly not great. However, by comparison, COVID-19 saw the world lose 255 million full-time jobs, or 8.8% of global working hours, according to the Sustainable Development Goal indicators website. Way worse, and we still don’t see mountains of white-collar workers living in refrigerator boxes, do we?
The more critical indicators are what AI has been doing to the job market in the last year and what it’s doing now. Are hundreds of corporate boardrooms stuffed with slavering CEOs looking to dump flesh-and-blood employees in favor of AI?
Rohit Patel, Director of Meta’s Superintelligence Labs, doesn’t think so. Meta Superintelligence Labs is where some of the most cutting-edge AI cookery is happening; he’s in a unique position to comment.
He believes the negative impact of AI on jobs is being overhyped. “It’ll have an impact for sure,” he says, “but it’s not going to be as damaging as is being predicted, and it’s not going to happen immediately.”
Reason being, and Patel sums it up nicely, a job is more than just a task, and individual tasks are all AI is capable of right now. “Because AI is relegated to tasks means it can help make the people who do those tasks more efficient, but that’s all you can reasonably expect right now. That’s why I don’t see any sort of job being replaced wholesale shortly.”
Patel believes that CEOs who focus solely on job consolidation are making a mistake. “A task isn’t a job because nothing is guaranteed,” he says. “A high performer today without AI isn’t necessarily going to be a high performer tomorrow with AI. To me, it’s less about cutting say 40% of your workforce right now, and more about taking the time to make a department more efficient with AI and then deciding whether that department needs as many people.”
AI in the real world
I spoke with Jim Yu, CEO of BrightEdge, a well-known San Mateo-based provider of SEO solutions. BrightEdge has been experimenting with AI for several years, but doesn’t sell an AI solution, so they’re a good example of what a tech-savvy business can do with this technology today.
Because SEO on a global scale requires massive amounts of data analysis that customers expect at an ever-faster rate, Yu says BrightEdge has been testing new technologies to help meet this challenge for some time.
“We started down this path about 10 years ago,” he explains. “That’s when our customers went from wanting the data so they could create their own insights to demanding that the app do it for them. So we started testing different technologies to help evolve the solution that way. We tested a lot of different approaches, and we had a lot of failures. But that’s the kind of experimentation and testing it takes to make the most of something new.”
To BrightEdge, generative AI has been a game changer, not just across application development, but also in the front office, particularly sales. Even so, when asked if he was considering laying off employees and replacing them with AI, Yu had a pretty strong response that jibes with Patel’s.
“No. And for those that are thinking that way, they’re going to be disappointed. AI can only do so much.” He also points out a key problem that most pundits aren’t talking about: You need to hold AI’s hand—a lot.
“What you’re looking for with AI is to make your employees more efficient and accurate, right? But that’s not plug and play at all,” he warns. “For us, it’s requiring endless tuning and testing across all kinds of use cases. It’s not some kind of magical solution. Getting to an 80% correct answer on an individual basis and for pretty basic tasks isn’t super hard, yeah. But what does that really get you? And getting beyond that 80% isn’t at all easy.”
So is anyone really on the chopping block?
Both Patel and Yu acknowledge that some jobs are at risk right now or in the near future, but neither one describes those consequences as a “white-collar bloodbath.”
“Can you use AI to replace someone right now?” muses Yu. “Sure, if you’re using people for basic research, for example. But that’s stuff like college interns. If you’re using three interns right now, you may only need one if your AI testing is successful enough.” However, he stresses that you’ll need to test AI rigorously before making that sort of decision, and that you’ll still need one intern to verify the AI’s results.
Patel is a little more expensive. “Beyond research, AI will definitely impact creatives fairly soon,” he predicts. “Particularly those involved in mass producing mediocre stuff. Like say you’re someone who creates the pictures that hang in hotel guestrooms. AI can certainly churn those out faster than people and the quality will be the same. That kind of job is definitely in jeopardy.”
But in either of these examples, even the creative one, AI still can’t just be turned loose to work on its own. Because AI is not reliable in any task. It’s often wrong without knowing it, yet still offers up those answers with authority. It also fabricates information, which Patel and Yu refer to as “AI hallucinations.”
“AI essentially works on an auto-complete model,” explains Patel. “It’s got a fixed data set to work with, and it’s just trying to answer your question with the information it has. It doesn’t have a concept of correct and wrong. That’s why sometimes you’re reading an answer and you’re asking, ‘Hey why did it just make this whole thing up’ Actually, I’m surprised it doesn’t do that more often.”
So an AI might turn up a wrong result that an intern wouldn’t have and deliver it with enough confidence that a reader might simply accept it if they didn’t take the time to check it out first. Or a creative AI might decide to simply reproduce an existing painting if it couldn’t come up with a variation of its own.
Examples of these kinds of hallucinations have already gotten some early AI users in trouble. For example, Forbes reported on a 2023 legal case involving Avianca Airlines and an irate passenger. The passenger’s lawyer was nearly sanctioned by the judge for using citations recommended by ChatGPT, which were found to be entirely fictitious, created solely to establish non-existent trials, quotes, and opinions.
However, when used carefully and with oversight, AI can have a profoundly positive impact on any business. Did I use AI when writing this article? Sure. Would my editor have fired me for just handing in a draft by [insert your favorite AI here]? Definitely, and it wouldn’t have been difficult for him to figure it out either. But when it came to finding all the market statistics and obscure punditry I cited earlier, it was invaluable and saved me at least a full day of work. So more efficient, but still in need of a human.
In Yu’s case, too, it’s been a game-changer. “The data we analyze is huge and also very individual to every customer. Analyzing data at scale is never going to be as efficiently done by a human as it is by an AI. So, it’s been phenomenal in that use case.”
That’s been impactful enough for him to recommend thinking about AI beyond just technology scenarios. “For instance, for us it’s been a huge help with sales simulators.”
Like most large companies, BrightEdge seeks to enhance sales performance by having every salesperson work through simulated scenarios. “But a busy sales manager can do that, what, once or twice a week for an employee? But with AI, we just trained it on the scenario and the customer’s sales and purchasing data, and now it can run those simulations much more often and also score the salesperson so their manager knows how they did.”
So if Patel and Yu’s experiences are accurate, AI will hurt the job market, but much more slowly and certainly less pervasively than is currently being predicted.
However, it will also create new roles even as it eliminates old ones. Just remember the WEF’s prediction above—AI might kill 83 million jobs in the next few years, but it’s also going to create 69 million new ones.
“Sure, there are going to be new jobs created by AI; it’s already happening,” says Patel. “In our labs today, we have PhDs who are creating new AI models and also building new LLMs on which our AIs need to be trained. That kind of job might be rare today and conducted only by experts, but as AI technology becomes democratized, businesses are going to need the same kind of skill set, which means they’ll have to hire people to fill this new role.” He also mentions that those new roles will need new software tools, which means you’ll see entirely new businesses being created, too.
When asked the same question, Yu cites BrightEdge’s recent experiences. “AI isn’t just changing what our software does; it’s changing how our developers are building it. That means in the future, we’ll almost certainly start hiring programmers who understand how to leverage AI, find the right AI, tweak and teach it, and design proper tests to vet it.”
For those who are sweating it
However, that might not mean much to worried white-collar and executive workers today.
Kelly Reeves, who is CEO of an executive coaching firm and host of the Confidence Academy Podcast, isn’t as bullish as Patel or Yu. But her blunt advice to those looking for new jobs or worried about keeping the one they have now still generally agrees with Patel and Yu’s opinions. She’s just not as rosy about it.
“A year and a half ago, AI was revolutionary,” she says, “but today, it’s a reckoning. If they want to succeed in the future, business leaders need to partner with AI and make it an integral part of how they manage and strategize.”
She’s particularly worried about the IT sector. “There’s already a drought of qualified IT candidates, which means in the short-term, there’ll be even fewer candidates once you factor in AI literacy.”
In the end, all three of my interviewees offered similar advice for job seekers and job keepers: you need to partner with AI and focus on how your own skill set can evolve with the technology, rather than trying to outdo it. Patel and Yu emphasize the importance of critical thinking and problem-solving skills.
“I’m telling my daughter she can major in whatever she wants, but she should also minor in data management,” says Yu. “Being able to solve problems and think critically is beyond AI, so focusing on disciplines that stress those skills is where I think a college grad should focus.”
Reeves agrees, but also thinks that those about to enter college look for programs to prepare them for jobs that don’t yet exist. “We’re entering an economy where we’ll need skills that aren’t being taught in most schools right now: AI workflow architects, prompt designers, cognitive experience designers, and AI compliance officers. Even things like AI cybersecurity skills are still evolving.”
Unfortunately, Reeves also believes that college students and professionals alike may have to design their own skill training to remain competitive, at least for now. “Curriculums simply aren’t keeping pace with AI right now,” she says. That means you’ll need to look to your own resources when it comes to living and growing with AI.
“Just start using it,” counsels Yu. “We’re encouraging all our people to see how an AI, even different AIs, can help them work better. We want them to think about jobs changing, not jobs being eliminated. That really doesn’t help anyone.”
Patel had even broader advice for the world at large: “If you’re worried about losing your job, you’re focusing on the wrong thing. You should be thinking about opportunities right now, not the negatives.” He cites the mid-nineties:
“Back when the internet suddenly exploded, there were lots of opportunities because nothing was online yet. Smart people were leveraging every opportunity to bring new things to the web,” he explains. “The same thing is happening today. For lack of a better word, the world is stupid right now. Figuring out how to bring artificial intelligence to any task or service is a fantastic opportunity that isn’t running dry anytime soon.”