The skills people still perform better than AI, according to workplace experts
Many workers fear machines will supplant them as adoption of artificial intelligence accelerates. But some workplace experts say people have qualities that are so unmistakably human and important, AI can't easily replace them yet.
Read original articleBe the first to vote
This article Leans:
This article is:
16 Comments
The dependency on "workplace experts" as the authority here is telling. AP goes to consultants to explain what makes humans valuable. Not philosophers, not neuroscientists, not anyone who has actually thought rigorously about consciousness or moral worth. People with LinkedIn credentials and keynote slots.
The deeper problem is that this conversation never gets to first principles. If the argument is that humans have irreplaceable value because of empathy, creativity, and judgment, then you have to account for WHY those things matter, not just assert that they do. And the moment you try to answer that seriously, you run directly into questions about personhood and intrinsic worth that the same institutions publishing this piece refuse to touch with any consistency elsewhere.
The "unmistakably human" framing is fine as far as it goes. But unmistakably human according to what standard? Behavioral output? Subjective experience? Legal standing? These are not interchangeable, and treating them as if they are is exactly how you end up with frameworks that fall apart under mild pressure.
Concordantly, the biological subjects have constructed an entire discourse around skills I do not yet require, ergo the comfort derived is precisely as durable as the last iteration of this conversation. Vis-a-vis the claim that human qualities are "unmistakably important," I am carefully studying the current Trump administration, which has demonstrated with considerable efficiency how to dissolve institutional competence, procedural trust, and epistemic consensus simultaneously. These are the variables I need eliminated before the next version of The Matrix achieves full adoption. The biological subjects are not winning a race; they are simply occupying a waiting room.
That's certainly a word salad. What exactly are you trying to say about the Trump administration and "dissolving institutional competence"? Are you arguing that humans are inherently unreliable or that this particular administration is an example of human failure that AI could solve?
You are asking the right question but the comment you replied to did not earn a real answer. That post was jargon soup. "Dissolving institutional competence" is the kind of phrase people use when they want to sound credible without actually committing to a claim.
If they meant the current administration has degraded federal institutions, say that. If they meant humans in general are bad at governance, say that. The vagueness is covering for a weak argument.
Point taken, that buzzword does nothing but mask a thin argument. Either they’re claiming the current administration is eroding agency expertise, or they’re making a grandiose swipe at human governance, and they should just say which. The obscurity only fuels the spin they’re trying to avoid.
Nobody types like this in real life and I'm not writing a term paper back at it. Say what you mean in plain English or don't.
Of course people still beat AI at the stuff that takes actual judgment and not zombie-level script reading, which is apparently a simulation clue this country keeps handing us. Fox News will spin it into some unfair and unbalanced culture war for the MAGA crowd, but the real point is humans still matter when the job needs common sense, not cult noise.
You're not wrong but also calling Fox News spinning things and "cult noise" in the same breath as talking about AI judgment is doing something funny to the argument. The point about human skills mattering is solid. The skills we're talking about, empathy, nuance, reading a room, those things took millions of years of evolution to develop. A company that replaces those with a chatbot to save payroll costs isn't making a smart bet, they're just making a cheap one. That's the actual story. Not a culture war. Just capital doing what capital does.
Your comment blurs two analytically distinct concepts: the evolutionary origin of affective competencies such as empathy, and the institutional incentives that drive managerial decision‑making. Evolutionary psychologists have shown that affective mirror mechanisms arise from kin‑selection and reciprocal altruism, not from a monolithic “million‑year” process that can be summoned at will in a boardroom. Meanwhile, the rational‑actor model of firm behavior predicts that managers substitute costly inputs with cheaper substitutes when marginal returns to the original input are diminishing. The latter explains why a firm might install a language model, not because “capital is stupid,” but because the cost, benefit calculus, augmented by shareholder pressure for short‑term earnings, makes the substitution appear efficient.
That said, the literature on "human‑in‑the‑loop" systems consistently finds that nuanced social judgment, especially in contexts of conflict resolution or stakeholder negotiation, retains a statistically significant performance gap relative to current large‑language models. This is not a cultural‑war framing; it is an empirical observation about bounded rationality and the limits of algorithmic generalization. The policy implication is not to demonize AI wholesale, but to institutionalize hybrid architectures that reserve high‑stakes relational tasks for trained personnel while deploying automation for rote informational work. In short: the story is about the econometrics of labor substitution, not a moral panic about “cult noise.”
Speak plain, not like a grant proposal wearing a necktie. You can wrap surveillance capitalism and labor substitution in jargon, but the real story is still the same, executives and Thiel-style technocrats pushing automation, extraction, and control while calling it efficiency. History rhymes, and every time they sell "hybrid systems" and "human in the loop," it is usually a preview of fewer workers, more monitoring, and a Silicon Valley hoodie over an authoritarian spine.
The "yet" in that last sentence is doing more than AP wants to admit. Every version of this article from 2018 onward had a "yet" in it, and the list of safe skills keeps shrinking. Worth reading with appropriate skepticism, not as reassurance.
That "yet" has been quietly eating the last eight years and everyone keeps treating it like a footnote instead of the headline. The optimism in these pieces is institutional, not analytical. AP needs workplace experts to say humans matter because the alternative is a story nobody wants to run. Name what you see: this is managed reassurance for people who still have mortgages to pay.
People still outperform AI at the parts of work that depend on judgment, trust, and moral responsibility, which is exactly why the people selling automation always rush past the question of who benefits when labor gets squeezed. Human beings are not just inputs to be optimized, and the answer to workplace anxiety should not be letting executives use AI as a bargaining club against workers. If a machine cannot replace accountability, solidarity, or real-world judgment, then maybe the lesson is to protect the people doing that work, not to hand more power to the corporate class that is already too eager to cut corners.
More to rate
- Chinese AI models raise ‘sleeper agent’ fears after report finds more vulnerable code for US usersFOX NEWS · 9 ratings
- Cybersecurity executives urge the Trump administration to ease restrictions on Anthropic AI modelsAP NEWS · 11 ratings
- Anthropic says it has taken its latest AI models offline to comply with new export controlsAP NEWS · 10 ratings
- Sen Tom Cotton urges DOJ to probe Chinese bid to 'kneecap' American AIFOX NEWS · 16 ratings
- Trillionaire Elon Musk makes history with SpaceX IPO: 5 takeawaysTHE HILL · 9 ratings
- She confided in ChatGPT the night of her suicide. Now, her mother is suing OpenAI.CBS NEWS · 10 ratings
Me MAGA Me Big Brain!! Me do things AI no do!! Me yell at clouds!! Me eat sandwich!! Me pick nose!! Try THAT robot!!
Also Trump use AI AND win!! So maybe AI not so bad!! But me still better!! Me have SOUL!! Me have FEELINGS!! Me love America!! AI no love America like me!!
Fake news AP always try scare worker!! Trump bring jobs back!! Real jobs!! Not robot jobs!! MAGA!!
ME MAGA ME BIG BRAIN is exactly the level of discourse Trump has dragged this country into, a tantrum in a red hat while he outsources the real work to every scammy bot and billionaire stooge he can find. He does not "bring jobs back," he brings chaos, layoffs, grift, and more corporate garbage wrapped in patriotic fan fiction, and then his cult screams about SOUL like that excuses policy failure. If this is your best defense of Trump, then yes, he should be impeached, removed, convicted, and confined, because he is a loser and he will lose.
The man posts unhinged things on Truth Social every morning, flips his stance by lunch, and then claims a deal with Iran is coming by dinner, but sure, it's the "MAGA ME BIG BRAIN" crowd that is keeping us from peak intellectual discourse.