Recently, Microsoft listed 40 careers that would be replaced by AI. Today, a friend sent me a link to a series of ChatGPT prompts where it could not follow a basic request to draw a human with two arms, two legs, and eight spider legs. It kept drawing ten spider legs, despite repeat corrections.
arxiv.org/pdf/2507.07935 I consider the paper and its methodology as fairly sketchy, but it doesn’t really say that: see section 4.3 In general they talk about careers being impacted, not necessarily replaced.
Fair; my language was taken from an essay that characterized the list as "careers AI will make obsolete." The authors say we can't know for sure what will happen in terms of loss of jobs or wages, but I do think the implication is there, regardless. www.independent.com/2025/08/20/g...
Agree on the implication. One of the aspects where the paper seems flawed. And presumably on purpose.
Doesn't really matter what the exact claims of that specific pre-print are. The overall point is valid: While tech "leaders" constantly make boastful predictions about replacing jobs, AI struggles to produce valid results for even basic tasks 🤷♂️.
Anyone who has ever used Outlook’s search function knows Microsoft software ain’t replacing anything
As long as nobody got fired for buying Microsoft...
Not so sure... Do you know "bad money replaces good money" theory?
Yeah. It’s not ever a substitute for a human. Although perhaps something inside of it may have a sense of humor… 😉
I don't believe any LLM will transcend the "uncanny valley" of human imitation. Language is essential to the human mind but doesn't embody it entirely.
Welp, my job as Nightmare Fuel Designer is safe for now
AI assistant on a phone call misspelled my name 4 times. I slowly spelled it letter by letter. WRONG all 4 times. It kept adding an extra letter.
Worse do it again a week later and get wildly different results
Microsoft has to keep saying that. How else will they stop losing money? This is on us for doubting them.
Ugh. Another consideration in this jobs elimination future…who gets promoted if entry level jobs are all AI? Who knows enough math & coding & geography & history & science & law, etc to ensure accuracy? If all learning and entry level jobs are outsourced.
One of the careers that Microsoft thinks will be replaced by AI is "geographer" and I really need us all to stop giving any credibility to tech bros, because someone definitely does not know what a geographer does if they think AI is going to replace us.
Pretty sure Geographers get additional magic powers based on the terrain type they stand on. Everyone knows that.
Yes that is all of the problem. People without any knowledge thinking they can create a machine that will have all the knowledge.
Lol. If you asked 20 geographers to define geography, you'd get 20 different answers.
Maybe! Most of us say it's anything to do with space and place (my MS and PhD are in geography).
Same here. The point I was clumsily trying to make is that AI can never replace a discipline that is so varied. Side note: I made the mistake of discussing "place" during my comprehensive exams and it opened the door to a lengthy discussion on concepts of place that put me way out of my comfort zone
no one ever suggests how ai can replace this
💯💯 too many US folks don’t understand the critical interdisciplinary and integrative work that geographers do, which AI will never be able to replicate. If they continue to assume it’s only cartography, even then AI will reproduce existing maps at best or makeup junk & make mistakes.
Amen!!!
Another geographer’s perspective on why Microsoft is wrong. www.independent.com/2025/08/20/g...
That’s the essay I initially read. In fact! As a few people note in the replies to my post, it turns out it misrepresents what the preprint actually said (the author may have been responding to the viral discourse before reading it). I agree with all the points raised, though!
The author should get the pre-print published elsewhere I think, so her position is clear. I don’t think all geographers who use AI are in the spatial sciences but many who are love using it, including physical geographers. I think critical human geographers likely have fought the AI trend longest.
They don't know what having a PhD entails (hint: It's not being an accurate answer machine you can put in your pocket) & they don't know what the actual practice of these professions is, and they can't be arsed to find out. Empty and uncurious.
I live with two geographers, and yeah ... that's a clueless thing to say
Microsoft said those careers would be affected by AI, as in AI would be used by people in those careers. Geographers already use machine learning to analyze data. www.esri.com/en-us/geospa...
I have a PhD in geography. I’m aware.
Yeah and you should know better and read the paper.
I did read the paper (as I note in the comments). There is very much an implication that some jobs could be replaced by AI, which is acknowledged by the authors in the paper. And certainly that’s how it’s being widely interpreted, which is understandable.
The way they measured was by looking at who was using AI for what. So it's just which careers will incorporate AI. The side effect might be needing fewer people to do those jobs, or it could actually mean more people will be needed to do those jobs because there will be more data.
I found out about the paper because a colleague at ESRI posted an essay by a UCSB geographer who described it as a list of jobs that AI would make obsolete. That’s not actually Ccurate, which I learned afterwards (to my frustration, since I trust the people involved to represent this accurately).
You’re right, but they don’t know what *any* of the people they’re firing to replace with unready AI actually do. The disrespect comes from ignorance, and they won’t realize what they broke for years, if ever.
This is a risk, but most businesses will recognize the before firing people. Federal govt, on the other hand... it seems to be" fire, ready, aim" over there since 2025.
Most AI (meaning not those machine learning programs designed properly but narrowly) are complete crap at geography, mixing up places, cultures, languages, environmental risks, etc.- let alone knowing how they are connected
I saw someone from MS the other night talk about using AI for early warning of disasters, and he was just doing linear extrapolation. He had no idea what needed to be done
I'm a GIS analyst and checked in on Copilot last week by asking it to make a map that highlighted every state with a cardinal direction in it's name. The results make it clear it's definitely not ready to take my place.
Never mind the fact that I REFUSE TO BE REPLACED.
✊🏽
Google’s AI summary told me just the other day that San Juan Island is not farther north than Maine, because Maine is further east. 🫡 🗺️ 💯 (It’s wrong.)
It's answer now is better, but still wrong (or at least severely misleading). All of San Juan Island is further north than all of Maine, but Google's AI summary says, "San Juan Island is further north than most of Maine." The sources it uses to back this up are two Instagram posts.
I think it’s just that it’s inconsistent more than anything. I don’t think there’s a been an update since a couple of days ago.
Good point. I'm glad for the example. I've used some of these with my students to explain why they shouldn't rely on the AI summary. My favorite was googling "difference between a sauce and a dressing." Answer: "sauces add flavor and texture to dishes, while dressings are used to protect wounds"
Google AI lies often
Especially when you ask chatgpt, and tell it the correct answer, it tells you the truth. KI is nothing for schools!
Oh, I know -- that's why I use Kagi.
Google’s AI summary has merged @leahstokes.bsky.social and I into a single person 😂
Oh my god! Too funny 😂
I'm not going to play with it, but I wonder how it answers "Can rivers flow north?"
Likewise, "Can mountain ranges run E-W?" (Incidentally, most of Oregon's internal rivers run north to north-trending.)
I can haz geographicalese!
Correction: this is not what Microsoft actually stated. I learned about the preprint from a friend at ESRI who shared an essay by a UCSB geographer who described it as “a list of jobs AI will make obsolete.” This wasn’t fact-checked by the paper, the professor, or my friend.
I get it. It’s going viral. I’m normally careful at fact-checking, but in this case I saw a published piece by a trusted source, shared by another trusted source. That was my first encounter with the Microsoft preprint. The preprint is a list of jobs with high AI “applicability.”
My vote is to pull the plug on AI. It’s too dangerous right now, especially with the felon in chief. Maybe one day when we have a genuinely altruistic governmental and political system…
A lot of people are interpreting this as “jobs AI can replace.” The authors even note that it’s tempting to do this, but job trends are difficult to predict. And I do think it’s reasonable to be concerned that employers will see these as jobs where AI can replace people. It’s already happening.
This is a good reminder to always go to the source. But I also think the response to this preprint is a good indicator that there are large-scale conversations about AI and labor that need to happen. And those researching this need to be more careful about how their work will be interpreted.
Labor AND the economic implications. What will be the policy response to job replacement? Taxing AI work hours? Unfortunately this administration isn’t thinking about a healthy society.
Someone that is close to me has used AI for very technical medical advice because the medical advice they have been getting sucked. And because things have been improving the Dr. has taken a victory lap. And it reminds me of the bikers that like WayMo because the cars are not threatening.
Which is to say, life is complicated, and I hope my loved ones survive the bad medical system and bike-hostile roads, and that collective action responds to this crisis.
I think this gets at the heart of my argument and message to our students: generative AI is one tool in your toolbox. You can use it, but it doesn't replace all of the other tools you should also have and use, because not every question/problem is a nail.
To wit: when my dad was dying of cancer, I used ChatGPT to "translate" the scan results/notes from his Dr. I didn't assume it was a stand-in/replacement for the Dr simply because I could understand it better, but it helped me to know what questions to ask the Dr to better home in on what dad needed.
"Maine is further east and THEREFORE further north"?? I'm curious if their AI knows what the Dunning-Kruger effect is...
To be fair, it also won't replace the jobs of people that someone at Microsoft thinks are geographers (GIS techs, at a guess).
Trisalyn Nelson at UCSB published on this in the local paper: www.independent.com/2025/08/20/g...
That is indeed where I heard about it. It turns out she mischaracterized what the preprint says (see the comments) but I do think the implication is there.
fortune.com/2025/08/18/m...
What frightens me is when they say "X could be replaced by AI" they don't mean "AI could do the work of X" they mean "we could replace the work of X with incorrect, randomly generated filler, and no one would stop us". What would the side effects be? They don't care, they already cashed the check.
I have heard that MIT recently did a survey asking corporations about their uses of AI. Apparently 95% considered the AI projects to be a failure. I considered this excellent news because it suggests that not everyone in the world has completely lost their minds.
Unless "95% considered it a failure" means "but they're sure the next one will be a success!". Idk
Lol. I'm sure companies will keep trying to find a place for it that actually works. Repeated failure tends to lower expectations, so maybe they will find a niche for it eventually. Idk.
Like as a non geographer it seems more than anything like a threat. "Do you like having accurate maps? We can take that away from you"
😳 have you actually read the paper?? The methods are just too precious 😬 I might have to write a thread or blog post about this, but I have to launch a new class in ten days, so it might take a while. Let's just say, I don't think geographers or historians should be too worried.
I read a response to the report by a geographer, but haven't checked out the report in detail yet. I'm sure it's very, uh, interesting!
interesting is a word you could use