According to a new study by researchers at Carnegie Mellon, MIT, Oxford, and UCLA,
Study should be solid I guess.
participants who were given AI assistants (in this case, a chatbot powered by OpenAI’s GPT-5 model) would have the aid pulled from them without warning during the test
Wow, interesting idea. 👍
where they had their assistant removed, the AI group saw the solve rate fall off a cliff. They had a solve rate about 20% lower
And even worse IMO:
They also had nearly double the skip rate, meaning they simply chose not to solve the questions.
This seems very alarming IMO, because this indicates they lost some of their ability to think constructively on how to actually solve a problem!
I know there have always been some who cried wold every time new technology has become available, like calculators and computers. Even dictionaries were once claimed to be harmful once!
But maybe this time there is a real danger, because AI takes away a lot of the need to actually think creatively and constructively. And that’s an ability we must not lose.
The last paragraph of the article is even worse. As it mentions 2 studies that show these effects are also long term!!!
When driving somewhere, if I set out with the mindset that I can’t rely on gps I can usually wing it and figure out where to go when a hiccup occurs. If I don’t, then I have a lot of trouble getting into that path finding mode when needed… similar to this maybe?
Yeah exactly, because although it’s possible to do more with technology sometimes, you’re actively de-skilling at the same time. When we invented the written word yes it legitimately made everything better, but also we lost oral traditions and the capacity to memorize large volumes of storytelling, songs, and histories. Now you can burn the books, and the knowledge dies. It’s a real risk.
Everything is like this. Every technology has a cost beyond its price, and making a decision of whether to use it or not will always be in error unless you think about what you’re losing in the process.
Changing the terms of the test in the middle of it, without warning, is disruptive. I’m not convinced it “fried their brains.” The same would happen with a calculator suddenly removed during the middle of an exam.
Here’s that last paragraph. Microsoft’s finding actually sounds like it does have the disruptive factor: people are trained to use AI and then it is removed. And finally, finally in the very last sentence of the entire article we get the one piece of information that’s been missing the entire time: doctors perform better with AI help, but then worse than ever without it.
My conclusion? Let people have AI and perform better with it.
Carpenters trained on power tools will suddenly perform worse with hand tools than carpenters who were never given power tools. But if they are given power tools, they can build homes faster.
No shit?
The findings are also in line with a study Microsoft published last yearthat looked at cognitive decline among knowledge workers, which found that the more people lean on AI, the worse they perform when asked to work without support. It also echoes a study out of Poland, which found that while doctors are better at spotting cancer risks with AI assistance, they perform worse than the no-AI baseline once that assistance is removed.
Carpenters trained on power tools will suddenly perform worse with hand tools than carpenters who were never given power tools.
Now you are just making shit up. None of these examples are about people being trained on AI. The comparison would be if a carpenter using power tolls for 10 minutes, suddenly becomes worse at using the traditional tools he is trained to use.
Your claim is baseless, there is no evidence for it, and the lack of any evidence of it, makes it an unreasonable assumption based on your prejudice alone, and should not be believed.
Let people have AI and perform better with it.
Again a very loaded statement, nobody is preventing anybody from using AI based on this research. But maybe people are not really performing better, or at least not always, it may depend on the task.
Your logic is fundamentally flawed and inconsistent, and you seem to lack any ability to see this as a potential problem, so much so that it reeks of you having an agenda.
Your flawed logic and prejudice does not beat 3 research papers.
Yes the article reporting on a research paper has an agenda, and not the random guy ignoring the evidence to contradict it. With absolutely zero to show for your argument, and clearly using flawed logic.
Ah yes, Gizmodo, arbiter of scientific truths. Their agenda is clear: to get you to click, typically with an outragey clickbait headline that reinforces your favorite narrative.
You need to learn the difference between debating someone and shouting at them that they have no argument, no logic, no evidence, and ill motivations. I can think of a couple other things you also need to do, but I’ll keep it PG.
Or any task change really. You tell me that I’m here for a writing task, then halfway through it becomes a math test? There’s no way I’m doing anywhere near as well as if they told me what was happening ahead of time.
If I use AI for my personal coding projects I’ve found that if the task is unsolvable by the ai model, I’m not able to sit down and do it myself until the next day. It’s like I’ve got to reset my brain.
If I want to save time and use AI for a specific part of the code, it probably saves me 5 hours of work. But then I spend five hours yelling at the ai to try to get it to actually solve it. Next day I’ll just fix it myself in 2 hours.
But what you’re describing is not that uncommon, even without AI: Oftentimes when trying to solve a complex problem and being unsuccessful you have to reset your brain by doing something fundamentally different or have a good night of sleep and after that you solve the problem easily.
May what you’re experiencing is not AI related at all.
You’re probably right, but I think it’s made worse by AI. Jumping into the code after 3 hours with Claude doing the dirty work feels like an impossibility
I’m really just tricking my brain to think I’m being more productive lmao.
But then again, some of the stuff I’m working on is in principle quite easy to do, but is also outside of my skillet, for these cases I benefit from using AI.
IMO the challenge is knowing how and when to use AI. Small companies using AI correctly can probably benefit massively from it. Although it’s risky
there have always been some who cried wold every time new technology has become available, like calculators and computers
and they kinda have a point, really. people got worse at memorizing stuff by heart when writing was invented, and people got worse at mental calculus when calculators when invented.
but they allowed many things that were simply not possible. a calculation that takes me 2 minutes in wolfram alpha could take hours if not days to solve by hand!
ai, meanwhile, or at least the ai we’re sold, does not offer significant advantages (at best it saves a few minutes), at the cost of making us worse at thinking, a skill that is absolutely essential to have… and of course, that’s the point. the tech oligarchs want us to be dependent on their extremely expensive products.
and people got worse at mental calculus when calculators when invented.
That may be true, but that is a much more limited problem, than losing some of our ability for critical thinking and problem solving in general.
ai, meanwhile, or at least the ai we’re sold, does not offer significant advantages
This is very true, the AI are shown to even hallucinate, and give incorrect and harmful solutions. A calculator does NOT do that.
So not only is the AI a danger to our critical thinking, we actually need it MORE when using AI.
No the test is not training, that’s a weird thing to claim. The switch is what is tested, and you disregard that 2 other tests have shown similar results. An actual decline in critical and problem solving thinking.
No the test is not training, that’s a weird thing to claim.
The control group solved 12 questions manually and then the 3 test questions manually. The AI grouped solved 0 questions manually and the 3 test questions manually. One group had 12 more manual math tasks to prepare for the manual math test the other group had 0 and also had to context switch.
The AI-assisted group was dealt a context switch, which results in a pretty severe performance loss. A context switch causes performance loss of around 40% according to this paper, which was peer-reviewed and published and is also the most cited paper on the topic, in the APA: https://www.apa.org/pubs/journals/releases/xhp274763.pdf
The AI-assisted group also did not have 12 questions to adjust to the new context, like the control group did. If they wanted to wipe out the context switching performance loss they should have kept asking questions to see if, after 12 questions, the AI-assisted group had a similar performance.
The switch is what is tested, and you disregard that 2 other tests have shown similar results.
No, they did not switch what was tested. Here is an image from the actual paper.
They were given 12 tasks with one group using AI and another doing mental math and then 3 tasks doing mental math. One group had 12 more tasks worth of preparation than the other.
Nothing, not even the article in theOP, says that they did math and swapped to reading to test.
They did 3 different experiments, in each experiment they gave 12 tasks and then disabled the AI for one group and gave 3 more tasks as a test. At no point did they ask 12 math questions and then finish with 3 reading questions or vice versa. They did 2 experiments using math tasks and 1 experiment using reading comprehension tasks.
So one group had 15 math tasks and one group had 12 ‘how to ask an AI’ tasks and then 3 math questions.
They also did not control for context switching losses, which is a well documented (see the APA paper) effect. The proper control would be to continue asking questions so the AI group also had 12 math tasks before the test.
There’s a reason that this is published on arXiv and not in a peer-reviewed journal. Designing a poor quality experiment doesn’t tell you anything useful even if you do multiple different versions of the same experiment.
This paper demonstrates a lack of a proper control group, specifically a failure to control for context switching performance loss.
The picture you post contradict your claims. The 2 groups are getting the same question, but one has AI assistance, the other has not.
Again you fail to show anything to support your claims.
The switch is what is being tested yes, but it is not clear that what is being measured in the switch is “AI fried their brains” rather than “context switching in the middle of a test”. If they wanted to make that point it would be useful to have the maths test run with a calculator group who also got it yanked halfway through, that way we would be able to see what proportion of the effect is over dependence on AI removing critical thinking and what amount is having your methods disrupted mid task.
The calculator test might be good for comparison, and I’m pretty sure if given the same amount of time, and one group being allowed to use calculator for half the test, that group would solidly outperform a group not using calculators at all.
I was in 5th grade in 1975, and we were the first class to get calculators in 5th grade. Which became the standard for many years after.
I have never heard complaints about students being less capable of understanding basic math problems because they use calculators. Although the idea of using calculators in schools were heavily debated. It’s similar to people not getting worse at spelling from using a dictionary.
The sessions lasted about 10 minutes, suggesting that those who decided to rely heavily on AI to solve problems for them abandoned their critical thinking abilities in a matter of minutes.
In academia it is normal not to directly spell out things that are obvious to a person with academic knowledge on the subject, research papers are meant for scholars, and they are supposed to be able to read and understand the consequences for themselves.
So you can’t use it as an argument that it isn’t spelled out, if it can be easily derived by a person who understands the subject.
Research papers do not spell out every possible consequence of their findings.
Study should be solid I guess.
Wow, interesting idea. 👍
And even worse IMO:
This seems very alarming IMO, because this indicates they lost some of their ability to think constructively on how to actually solve a problem!
I know there have always been some who cried wold every time new technology has become available, like calculators and computers. Even dictionaries were once claimed to be harmful once!
But maybe this time there is a real danger, because AI takes away a lot of the need to actually think creatively and constructively. And that’s an ability we must not lose.
The last paragraph of the article is even worse. As it mentions 2 studies that show these effects are also long term!!!
it has ruined the ability of K-12 people writing and reading proficiency.
When driving somewhere, if I set out with the mindset that I can’t rely on gps I can usually wing it and figure out where to go when a hiccup occurs. If I don’t, then I have a lot of trouble getting into that path finding mode when needed… similar to this maybe?
Yeah exactly, because although it’s possible to do more with technology sometimes, you’re actively de-skilling at the same time. When we invented the written word yes it legitimately made everything better, but also we lost oral traditions and the capacity to memorize large volumes of storytelling, songs, and histories. Now you can burn the books, and the knowledge dies. It’s a real risk.
Everything is like this. Every technology has a cost beyond its price, and making a decision of whether to use it or not will always be in error unless you think about what you’re losing in the process.
Changing the terms of the test in the middle of it, without warning, is disruptive. I’m not convinced it “fried their brains.” The same would happen with a calculator suddenly removed during the middle of an exam.
You are disregarding the last paragraph, where 2 other studies showed similar results, without having the “disruptive” factor.
Here’s that last paragraph. Microsoft’s finding actually sounds like it does have the disruptive factor: people are trained to use AI and then it is removed. And finally, finally in the very last sentence of the entire article we get the one piece of information that’s been missing the entire time: doctors perform better with AI help, but then worse than ever without it.
My conclusion? Let people have AI and perform better with it.
Carpenters trained on power tools will suddenly perform worse with hand tools than carpenters who were never given power tools. But if they are given power tools, they can build homes faster.
No shit?
Now you are just making shit up. None of these examples are about people being trained on AI. The comparison would be if a carpenter using power tolls for 10 minutes, suddenly becomes worse at using the traditional tools he is trained to use.
Your claim is baseless, there is no evidence for it, and the lack of any evidence of it, makes it an unreasonable assumption based on your prejudice alone, and should not be believed.
Again a very loaded statement, nobody is preventing anybody from using AI based on this research. But maybe people are not really performing better, or at least not always, it may depend on the task.
Your logic is fundamentally flawed and inconsistent, and you seem to lack any ability to see this as a potential problem, so much so that it reeks of you having an agenda.
Your flawed logic and prejudice does not beat 3 research papers.
I laugh in your face. This article has a clear agenda, not me.
Yes the article reporting on a research paper has an agenda, and not the random guy ignoring the evidence to contradict it. With absolutely zero to show for your argument, and clearly using flawed logic.
All I hear is the laugh of ignorance.
Ah yes, Gizmodo, arbiter of scientific truths. Their agenda is clear: to get you to click, typically with an outragey clickbait headline that reinforces your favorite narrative.
You need to learn the difference between debating someone and shouting at them that they have no argument, no logic, no evidence, and ill motivations. I can think of a couple other things you also need to do, but I’ll keep it PG.
Ah yes the classic blaming the messenger argument, that is one of the most obvious and stupid fallacies.
Or any task change really. You tell me that I’m here for a writing task, then halfway through it becomes a math test? There’s no way I’m doing anywhere near as well as if they told me what was happening ahead of time.
If I use AI for my personal coding projects I’ve found that if the task is unsolvable by the ai model, I’m not able to sit down and do it myself until the next day. It’s like I’ve got to reset my brain.
If I want to save time and use AI for a specific part of the code, it probably saves me 5 hours of work. But then I spend five hours yelling at the ai to try to get it to actually solve it. Next day I’ll just fix it myself in 2 hours.
But what you’re describing is not that uncommon, even without AI: Oftentimes when trying to solve a complex problem and being unsuccessful you have to reset your brain by doing something fundamentally different or have a good night of sleep and after that you solve the problem easily.
May what you’re experiencing is not AI related at all.
You’re probably right, but I think it’s made worse by AI. Jumping into the code after 3 hours with Claude doing the dirty work feels like an impossibility
That sounds a lot like what the studies show. And IMO that sounds like a serious problem.
I’m really just tricking my brain to think I’m being more productive lmao.
But then again, some of the stuff I’m working on is in principle quite easy to do, but is also outside of my skillet, for these cases I benefit from using AI.
IMO the challenge is knowing how and when to use AI. Small companies using AI correctly can probably benefit massively from it. Although it’s risky
and they kinda have a point, really. people got worse at memorizing stuff by heart when writing was invented, and people got worse at mental calculus when calculators when invented.
but they allowed many things that were simply not possible. a calculation that takes me 2 minutes in wolfram alpha could take hours if not days to solve by hand!
ai, meanwhile, or at least the ai we’re sold, does not offer significant advantages (at best it saves a few minutes), at the cost of making us worse at thinking, a skill that is absolutely essential to have… and of course, that’s the point. the tech oligarchs want us to be dependent on their extremely expensive products.
That may be true, but that is a much more limited problem, than losing some of our ability for critical thinking and problem solving in general.
This is very true, the AI are shown to even hallucinate, and give incorrect and harmful solutions. A calculator does NOT do that.
So not only is the AI a danger to our critical thinking, we actually need it MORE when using AI.
But they’re using the hell out of it, too, right? They’re exactly the types of people that love and use it the most: managers and owners.
This paper shows that a person who has performed a task 12 times performs better than a person who has never performed the same task.
They also do not properly control for performance loss due to context switching which is a well known contributor to performance loss.
It’s a paper on arXiv, it hasn’t been peer reviewed or published.
No the test is not training, that’s a weird thing to claim. The switch is what is tested, and you disregard that 2 other tests have shown similar results. An actual decline in critical and problem solving thinking.
Here is the paper: https://ai-project-website.github.io/AI-assistance-reduces-persistence/
The control group solved 12 questions manually and then the 3 test questions manually. The AI grouped solved 0 questions manually and the 3 test questions manually. One group had 12 more manual math tasks to prepare for the manual math test the other group had 0 and also had to context switch.
The AI-assisted group was dealt a context switch, which results in a pretty severe performance loss. A context switch causes performance loss of around 40% according to this paper, which was peer-reviewed and published and is also the most cited paper on the topic, in the APA: https://www.apa.org/pubs/journals/releases/xhp274763.pdf
The AI-assisted group also did not have 12 questions to adjust to the new context, like the control group did. If they wanted to wipe out the context switching performance loss they should have kept asking questions to see if, after 12 questions, the AI-assisted group had a similar performance.
No, they did not switch what was tested. Here is an image from the actual paper.
They were given 12 tasks with one group using AI and another doing mental math and then 3 tasks doing mental math. One group had 12 more tasks worth of preparation than the other.
Nothing, not even the article in theOP, says that they did math and swapped to reading to test.
They did 3 different experiments, in each experiment they gave 12 tasks and then disabled the AI for one group and gave 3 more tasks as a test. At no point did they ask 12 math questions and then finish with 3 reading questions or vice versa. They did 2 experiments using math tasks and 1 experiment using reading comprehension tasks.
So one group had 15 math tasks and one group had 12 ‘how to ask an AI’ tasks and then 3 math questions.
They also did not control for context switching losses, which is a well documented (see the APA paper) effect. The proper control would be to continue asking questions so the AI group also had 12 math tasks before the test.
There’s a reason that this is published on arXiv and not in a peer-reviewed journal. Designing a poor quality experiment doesn’t tell you anything useful even if you do multiple different versions of the same experiment.
This paper demonstrates a lack of a proper control group, specifically a failure to control for context switching performance loss.
The picture you post contradict your claims. The 2 groups are getting the same question, but one has AI assistance, the other has not.
Again you fail to show anything to support your claims.
I also wrote text.
If you’re just going to cherry pick a single point and dismiss everything else then we’re done here.
The switch is what is being tested yes, but it is not clear that what is being measured in the switch is “AI fried their brains” rather than “context switching in the middle of a test”. If they wanted to make that point it would be useful to have the maths test run with a calculator group who also got it yanked halfway through, that way we would be able to see what proportion of the effect is over dependence on AI removing critical thinking and what amount is having your methods disrupted mid task.
The calculator test might be good for comparison, and I’m pretty sure if given the same amount of time, and one group being allowed to use calculator for half the test, that group would solidly outperform a group not using calculators at all.
I was in 5th grade in 1975, and we were the first class to get calculators in 5th grade. Which became the standard for many years after.
I have never heard complaints about students being less capable of understanding basic math problems because they use calculators. Although the idea of using calculators in schools were heavily debated. It’s similar to people not getting worse at spelling from using a dictionary.
Not training, no, but warm up. And no, it is not about critical thinking, it’s about reading comprehension and calculations.
Also and this is the big one for me. It’s 10% wrong on average. That’s really bad. 1 in 10 google Gemini answers is bullshit
And the ability to think critically to detect it declines. So it’s doubly harmful!
1 in 5 human answers is probably bullshit so it sounds like you’re onto a winner
Wow. Now do this with a calculator.
A calculator is not the same problem, it doesn’t reduce our general ability to think critically.
The studies referenced are about calculations, reading comprehension and work performance, not critical thinking.
The article is, like many, a bad one. It generalises what it should not.
As I said, this is a bad article. The experiment does not suggest that at all. The study does not mention critical thinking.
I’d say, however, that the proliferation of shitty news websites has caused readers to lose their critical thinking.
In academia it is normal not to directly spell out things that are obvious to a person with academic knowledge on the subject, research papers are meant for scholars, and they are supposed to be able to read and understand the consequences for themselves.
So you can’t use it as an argument that it isn’t spelled out, if it can be easily derived by a person who understands the subject.
Research papers do not spell out every possible consequence of their findings.
It isn’t spelled out because it is not a logical conclusion at all. Nothing in this test requires critical thinking to achieve.
Why are you defending an obviously terribly written article?
As the study defines critical thinking: yes it does.
The study claims, essentially, relying on a machine that solves a Problem for you, lessens your critical thinking skills.
Their Definition of “critical thinking” is just, at least to me, way Off.
Just because i can conprehend Stuff i read for example, does not show critical thinking. It just shows i can repeat shit i read adequately.
It’s just bad science.