Simo Hosio
4K posts
Simo Hosio
@simohosio
Optimist Professor in Computer Science & Engineering / 🇫🇮 & 🇯🇵 / Turbocharge your academic output 👉 https://t.co/GagGgjuNGi
Click me: Katılım Ocak 2017
578 Takip Edilen3.7K Takipçiler
PSA: Please, for the love of science, do NOT use AI to generate your paper figures. Regardless of the quality of the underlying paper, an AI-generated diagram conveys sloppy thinking and causes readers to mistrust your research entirely.
English
Students are just like all other human beings, happily blaming everyone else for their own cheating
We see this in our studies in the mature population as well. Nothing is ever your own fault and you have no agency over anything ( okay okay, I'm exaggerating but that's the message we get from all humans not just students. )
English
Not that my experience is representative, but anecdotally students I know well are overwhelmingly negative about AI with some saying explicitly that it prevented them from having the college experience they’d imagined.
kang@jaycaspiankang
one thing I noted was how many professors talked about how the students aren't all thrilled to be using AI, even as they use it. There is a question: what is the point if we can do this?
English
@GroksBride Back in the 60s we didn't really even have all the ingredients to bake similar cakes as today.
I really do like cottage cheese though
English
@simohosio So if I gave you an investment in cash for a well structured tree house for you and your friends and a case of beer, can you return my investment 10 fold?
English
Dear Graduate Universities. Explain to me how you can charge $50k–$100k for something students can now learn themselves? And when they do partake in academia and seek non-AI assistance from a professor, the professor’s response is to ask if they reviewed the class PDF or video? Please justify your existence. I’ll wait.
English
At this point I can vomit out a paper in a weekend. A bit later everyone can. But if you want to be able to discuss and defend your work and actually understand what you created, you better linger on it a bit longer.
Studying the work you cite will be the mental gym for those who still want to understand instead of just manifest papers.
What a gorgeous time to be alive
English
@kkthomason Hahahah, i sometimes make posts on how great it's to be an academic. The rage I get :D
English
Is there another profession as self-hating as academia? It's shocking to me how many of my colleagues in the profession fall all over themselves to heap scorn on everything we do.
English
Not a bad idea to just record your wins daily. If you focus all the time on the inevitable work ahead (to hit your targets), your head will be in struggle mode nonstop.
Just notice what goes well in your life.
Pic very much related, top reply to "what's the one habit that genuinely improved your life" on Reddit this month.
English
The number of papers, as a proxy for doing our job, has been a disaster for academia.
And now, we pay the price.
English
Don't do this for camera ready papers.
Do this for submissions.
And if you even TRY to submit hallucinated BS, you just get locked out of the system for a year.
We should take no prisoners.
English
@rammy_c8 because in the right place it's easily the most fun time of your life?
English
Not the easiest job in the world + raw talent alone is not even enough. Plus luck. None of these attempts to model "how to make it" can factor in the absurd role sheer luck has.
(Bornmann, L. (2024). Skewed distributions of scientists’ productivity: a research program for the empirical analysis. Scientometrics, 129, 2455–2468.)
English
@timgill924 Four weeks is quite the luxury. A more typical scenario is something like five days of time after nine months of waiting for the decision
English
@Haliwa_Saponi The person who writes a paper obviously with AI has to make the time to explain themselves or just face the consequences
English
Asking a student to make time to come to office hrs to tell you what a paper is about (when they don’t need or want your help) is nuts when you could do your job and just read the paper…
This is why so many ppl are turning away from traditional school. Who has the time?
Dr. Mia Brett@QueenMab87
Asking a student what their paper is about is not a whole other assignment. It’s part of helping students with their work. It’s literally part of teaching. And oral exams teach completely different skills. But I guess you could choose failure instead of a 5 min convo
English
@IMurtazashvili Professors don't want articles to be the currency of academia btway. There are powers above professors who have decided so.
English
If students can get away without reading a book, maybe it is because of the vast number of professors who think we don't need to read books and "articles are the currency of academia." It's a reflection of the academic publishing model, not something new with AI (the article should probably be titles you can get a degree without reading a book or anything else - that's the issue with AI, it's not about reading books per se).
The Telegraph@Telegraph
An Oxford academic has warned that students using AI can obtain a degree without reading any books Katherine Rundell warns that reliance on AI is creating a ‘vast counterfeiting of knowledge’ in universities 👇 telegraph.co.uk/news/2026/05/2…
English
Personally I'd be pretty okay with that. Cannot possibly imagine a situation where even one fabricated citation would be okay or forgivable
Just the discussion here has been absolutely bonkers and completely unreasonable but then again I wouldn't expect anything less on this platform
English
@simohosio @nataliemj10 That's disingenuous. He's not being asked to read every paper. He's being told that all authors will be held responsible (quite reasonably IMO) if AI-hallucinated citations are included in a manuscript. How a research team handles that is up to them.
English
Granted, I'm not in academia and haven't been since a postdoc 15 years ago, but...
It honestly never occurred to me that an author *wouldn't* be held responsible for the existence and accuracy of every citation, ESPECIALLY in academic work. You want to be taken seriously, right?
James Miller@JimDMiller
So this means you expect every author to check every citation and make sure that every citation is real and accurate? What if it's beyond the ability of one of the authors to verify one of the citations because that citation is in a language he doesn't know or concerns technical material he doesn't understand but another author on the paper does?
English
@timurkuran Well the alternative is that we need students that are not cheating
And the ones who cheat hopefully understand that they will graduate with no skills
English
AI just killed higher education’s old teaching model. We need smaller classes and oral defenses for every paper—implying more faculty time, hence more professors. Since banning AI is unenforceable, written work alone can no longer be trusted.
English
AI detectors were a quick business model but I think they should die. More trouble than gain. Ridiculously unreliable. Plus if you are not an absurdly lazy human being, you can simply use your AI on a second screen and then just talk your stuff into existence on the other one using a microphone. In your own voice and tone.
Burn them all
English
English
@GNOBOTS @jimstewartson oh god.... lol....
i should give you another chance to read...
neural networks are not a lookup database as Jim suggested.
English
If you believe that LLMs are querying a database for answers
you're decidely unqualified for this discussion.
Jim Stewartson, Decelerationist 🇨🇦🇺🇦🇺🇸@jimstewartson
I’m getting tired of “experts” like this misunderstanding what they’re looking at. LLMs are giant databases of stuff HUMAN BEINGS have done. They are the EXHAUST of humanity. Prompts are database queries into EXISTING DATA. It’s a fuzzy search engine, not intelligence.
English
@CoreyMiles__ Liking and being forced to are two different things 🤷
English
do academics even like research, reading, and thinking or just publishing
Serafim Batzoglou@s_batzoglou
@michaelbilleaux To be a successful PI in bio it usually (not always) requires writing 10-20 high profile papers a year, each one with ~80 citations. No bio PI reads each one of these citations cover to cover. That would be absurd.
English
