> In a listening session with instructors, we learned that some were considering using AI agents as research assistants instead of hiring undergraduates as UROPs.
If researchers at a well funded institution like MIT are seriously considering replacing hiring undergrads with LLMs, I can just imagine how researchers at schools with less funding are open to it. I never did research as part of my undergrad (something I regret, but I likely wasn’t in the headspace for it back then) but my friends who did view it as a core part of their education and deeply helpful for future opportunities. I really hope schools come up with a policy to help discourage this.
cs_throwaway
If you know how to use an agent, and I do, you can have do in minutes what would take a undergrad all summer to do.
But undergrad research has always served two roles: getting my work done and the pleasure of giving the best students the chance to go beyond the curriculum, which is designed for the average student.
I think the impact of AI on undergrad research opportunities will be zero. The bigger impact is reduced science funding. I have directed lots of PhD student grant funding to undergrads who deserved it. I don’t see myself doing that in the near future.
us-merul
I think this has been going on well before AI or LLMs: "A fundamental danger, as we’ve discussed, is that AI can allow students to bypass learning. Equally concerning is that students may internalize a transactional model in which assignments are outputs, teachers are evaluators, peers are optional, and knowledge (or an MIT degree) is an optimizable commodity to be acquired or produced as efficiently as possible."
miltonlost
Ok, but did AI and LLMs make the cheating easier, faster, harder to detect? Whatabout the past doesn't change the problems now.
"My leg already had a cut on it. No need to worry about the new stab wound in my chest!"
us-merul
Absolutely; it's a real problem. But institutions like MIT have also been participating in, and benefiting from, this transactional relationship from their side in offering degrees that they know carry so much value.
bloppe
MIT makes all its money from research (and investing its endowment more successfully than most schools), not tuition, and the success in research comes from having such a high-achieving community of faculty and graduate students. They're pretty incentivized to keep the bar for academic achievement genuinely high.
us-merul
To clarify, I am pro-university. I'm just saying that institutions do not always behave in ways that are purely for educational benefit, such as with legacy admissions or grade inflation. I don't think the blame falls solely on students for viewing university degrees as a necessary step for something else. I would sure bet admissions committees will use AI to screen applicants, if they aren't doing so already.
alex_c
I'm surprised at the negative comments here. I'm about halfway done reading the report - I think it's clear, well written, and quite frankly the opposite of fluff.
A document like this isn't going to magically "solve" the use of AI in higher education. The purpose is to define a shared understanding of the situation across a large, complex organization, and set an initial direction and general shape for actions to take.
It contains clear, specific observations of how AI is organically changing the reality of education. And, in my opinion, fairly clear high-level guidance on what MIT as an entity wants to do about AI, and what individual departments and faculty should decide on their own.
A document like this doesn't need to be revolutionary, it may feel like fluff because no specific item in it is particularly surprising or groundbreaking, but the value is in having the entire document as a whole. And it is a lot more comprehensive and well thought out than what most companies can put out.
losvedir
I think you can tell who still has the ability to read long form content by whether they thought this was full of fluff or decently actionable. I think the bolded parts of section 3 are definitely not fluff. I found the document pretty interesting.
phasefactor
It can be both full of fluff and have actionable take aways.
This is dozens of pages long and reading it thoroughly will net you maybe a dozen useful sentences. That ratio is the problem people are complaining about.
21wa
Yes, only MIB (Massachusetts Institute of Bureaucracy) alumni have the capacity of understanding such an august publication!
UJ-FLD-ScrTr
You can think what you like about MIT, but at least they're giving it some thought, however perfunctory it may seem. Many German universities prefer to try and ignore the issue and hope for the best.
CalRobert
I mostly think about what they did to Aaron Swartz.
123h-asdg
They also formed committees and investigated after the fact back then!
frognumber
And then they adopted it as a tactic.
No, I can't provide a cite. Insider information. But whistleblowers have been threatened with being "Shwartzed" since (e.g. have frivolous criminal charges filed, to make their lives hell). In every case I've seen, whistleblowers signed an NDA, a non-disparage, and backed down.
That's why I don't trust a !@#$% thing from MIT anymore. Lies and fraud get covered up. Once that starts happening, you can't trust anything.
UJ-FLD-ScrTr
Regarding Swartz, I agree, but those are two different things. My statement was also more about the avoidance of the topic at German universities.
DataDive
This seems to have touched a nerve, based on the number of downvotes.
In this particular case that makes the message more trustworthy for me ;-)
ncr100
How is this interesting? Sincerely?
Tell us more about how MIT giving AI some thought compares superiorly to 'many German universities', and please tell us more about which German universities and in which circumstances.
UJ-FLD-ScrTr
My statement was more of a dig at German universities than a compliment to MIT. I understand your criticism of this institution (which doesn't even exist in that form in Germany), but that doesn't mean I have to reflexively condemn everything from there.
ncr100
Oh of course, I understand.
I'm just ignorant to German universities. Honestly, the only thing I know of is not even University level German education, but just the concept of the German gymnasium system, and that was from when a exchange student visited my high school, in the 1990s!
Thanks for the response. Cheers.
UJ-FLD-ScrTr
Then I hope you have better memories of our country.
To explain briefly:
Broadly speaking, there are these types of schools:
in various organizational forms. At a Gymnasium, you can obtain a full Abitur (university entrance qualification), which qualifies you to study at
- Universities
- Universities of Applied Sciences
For studies at Universities of Applied Sciences, a Fachabitur (subject-specific university entrance qualification) is necessary, which can be obtained in various ways. Previously, diplomas were usually awarded there, an academic degree between a B.A. and M.A. Since the Bologna reforms, the system has been changed to B.A./M.A.
At Universities of Applied Sciences, studies are more practically oriented (with practical phases); however, the biggest difference is that you cannot pursue a doctorate there, which is only possible at a "proper" university.
Besides social informatics, I also teach academic writing, which is why the topic of AI is important to me, because it's not easy for teachers to deal with. But I also don't think the strategy of simply looking away is effective.
testfoobar
This is a bunch of fluff. I hope at least the snacks and lunches during the discussions were good.
"Guiding principles: Be bold. Be humble. Put humanity front and center. Lean into learning. Teach with intentionality. No one size fits all."
"Recommendations: Adapt educational processes for an AI-aware world. Center people, community, and the residential experience. Build processes, teams and tools for continuous reflection, iteration, and improvement"
ChicagoBoy11
Yes, but as a K-12 administrator, I sadly have to tell you this is just about one of the more substantive things I have seen.
Yes, it's 95% fluff, but there is a real admission here that the guidance really ought to accept that there will be a whole host of tasks/assignments that kids engage in where the assumption SHOULD be that they basically will leverage AI, and that for their own benefit, resources should be set aside to promote experiences -- both in the context of assessments but also even in the social sphere -- there the influence of AI will be very purposefully prevented.
That is an actual stake in the ground, I think (as far as it goes in context like this). I think I would have been hard-pressed to have predicted that there would be such a revolutionary technology where the explicit ask of MIT faculty would be to prevent its influence in the school.
testfoobar
I really do worry for our younger students - and I don't envy your job at all.
It seems as if all of teaching and student evaluation needs to be rebuilt from the ground up with AI as a default assumption.
Local high schools have added in class timed, hand written, essays for classes like AP US History.
I particularly like Chicago Law School's policy on AI.
E.g. Law students writing a "Substantial Research Paper" will have to defend it orally: "We will be adding one additional requirement, which is that all students will be required to engage in an oral discussion of their SRP with their supervising professor, in an in-person setting. This discussion will occur after a complete draft (or final version) of the paper has been submitted to the professor. The discussion could take place one-on-one, or as a class presentation in the style of an academic workshop. Either way, the oral exchange will involve the student answering questions that probe the reasoning of the paper and the implications of its arguments."
It will only get harder. From the trenches in tech it’s hard enough to get people who should know better to put in the work to understand the ai output.
For education I can’t imagine. You might have to resort to actual discussion in classrooms instead of homework, with the accompanying need for both more educators and higher quality therein. Which isn’t going to be easy when education is fully under regulatory capture (at least for K-12 in the USA)
ghaff
>actual discussion in classrooms
Which works in some contexts; I've taken seminars.
Not sure how that works with a multi-hundred person core calculus, physics, or chemistry class.
lazyasciiart
They did say k-12.
wafflemaker
>K-12
K-12 means the public school system in USA & canada. It consists of kindergarten, elementary , middle and high schools, encompassing 12 grades/classes, each taking one year to complete.
fc417fc802
Maybe it just requires approaching the problem differently. What about the vulcan study pod things? Have the discussion with the AI itself. Have the AI regularly assess understanding.
ChicagoBoy11
So this is an alternative that I actually don't object to at all, but you still need to create mechanisms where you then get at the quality of the mental models formed without the assistance of AI. My (very unpopular) hot take is that some of this will actually usher in a new era of "standardized test"-like things; we will be far more permissive about how/in what ways students/professionals acquire knowledge, and the real task will now shift from that to guaranteeing that there is some competency that was reached with the assurance that it isn't masked by an outside influence.
If you basically assume that everyone can and will cheat at all times, then you can let yourself be far more agnostic about it AS LONG as you are confident in some ability to measure the person's skill without the tools.
plasticchris
The interesting thing is that ai actually helps me learn more effectively than ever before. Each session can become a personalized little seminar that dynamically tests me as it presents ideas and concepts and I attempt to apply and extend them.
I think we need to move past recall as the test itself, and tasks as the gate we judge academic achievement by. These were only ever an approximate measure, which is why we still have things like thesis defenses.
But imagine the final being an interview with an agent that is designed to probe your mastery? From my experience so far it seems we could measure much more directly the educational outcome and also accelerate the learning at the same time. But this assumes those involved want to learn, which widespread cheating suggests is not the case. Maybe the real answer is that we shouldn’t attempt to force curiosity but instead cultivate it where we find it, allowing those who reject it to be redirected until they find something engaging.
rawgabbit
I thought the recommendations in section 3 were substantiative. That is move away from time-bound exams and move towards semester length project portfolios.
>"In the era of AI, some traditional learning goals may merit rethinking; for example, do the majority of our students need to be able to write complex programs by hand?"
>"quick, high-stakes evaluations embody the opposite of the signal we want to convey to them right now."
>"We urge instructors to consider forms of assessment that are less vulnerable to AI, and more valuable for learning, such as oral exams, semester portfolios, and out-of-class assignments paired with in-class conversations. This likely means resources such as TAs and class time will become more central to evaluation."
ericpauley
Project-based learning is not a new idea. However, it is precisely the sort of assessment strategy that is being attacked by AI: students can complete many projects fully using AI without learning anything about the underlying material.
ghaff
In general, project/lab courses are pretty useƒul even iƒ they need some lecture/theory tacked on. Certainly, in school, I found hands-on and other project work, both individually and in groups, some of the better course work I did.
cptroot
I don't agree? By the time you get to the specific policy sections (3.x.x) there are concrete changes outlined in bold, several of which are requests for the institute to invest large amounts of resources changing the current teaching process. Even the non-bolded text occasionally contains real insights, like this gem from the end of 3.1.9:
> Students should not feel policed. Durable change will require instructors to be as clear as possible about their expectations and students to understand AI misuse as an unacceptable deviation from shared peer norms and community values rather than a violation of an arbitrary bureaucratic rule.
washadjeffmad
I immediately forwarded 3.3 to a colleague (AD of Research). We have a blanket policy document at the institutional level to establish a framework for baselines, but it's absent of any practicable guidance outside of a general approval process and expectations for academic honesty.
123h-asdg
They have always dressed it up as concerned humanitarianism. They did it already with large donations in the past:
All concerns, while the real goal is stated right in the article:
"With this new approach, he says, experts in many fields can gain a deeper understanding of AI so they can better harness it. Meanwhile, computing experts are gaining greater exposure to the work of their counterparts in other fields. That exposure is giving the technologists a better understanding of how to create and train AI tools to better serve others."
I'm sure they will have many meetings about future meetings.
frognumber
Makes me sick.
MIT is so very, very overcapitalized.
These things don't lead to anything other than a rich guy getting his name on a building, and more corruption in academia. By far the best thing which could happen to MIT would be for 95% of the endowment to go up in a poof of smoke, bringing it back to nineties levels. That's the end of the period when the Institute did high-integrity research.
jacobolus
It sounds like you want a return to a time before a lot of the government research money coming to institutions like MIT was gutted.
frognumber
No, today institutions like Georgia Tech -- which has an endowment of just about the same as MIT in the nineties adjusted for inflation -- seem to function just about the same way as MIT did back then.
I want to get back to the era before people went to MIT as a center of wealth, power, and prestige, and would lie and cheat to get tenured positions there, to where it was a place where nerds went to nerd together.
And to a place where every !@#$% done didn't lead to a self-hyping press release, but where statements were precise, measured, calibrated, and most importantly, scientifically-accurate.
plasticchris
Anywhere the nerds or artists go, if they are good, eventually becomes overrun with such types. Just look at the state of Silicon Valley and y combinator, and gentrification in general…
FWIW: I see this pattern each time I discuss MIT (drifting from the Institutional Propaganda partyline) on HN:
1. Upvotes
2. Time passes
3. Downvotes
I'm starting to suspect Institutional astroturf.
rpaik
It's going to be difficult not to be "fluffy" as no one really knows the true impact and what's going to happen. And it's especially challenging for universities as voices questioning the value of higher education will only grow louder.
BubbleRings
I grew up with PC‘s starting with DOS 1.0. I built a career as a systems engineer, basically by being confronted with thousands of problems that took four hours to figure out because the Windows world was changing so fast there was no documentation to easily find your answer. Those same problems now take me five minutes to figure out by using AI, so I’m able to move onto the next challenge and learn the next thing. That’s one aspect that I think isn’t well recognized in the anti-AI mindset.
That fact, multiplied over 1000 different disciplines, is one of the ways that AI is going to bring amazing progress to our world.
dpweb
They have to do something. There's a cheating epidemic right now everywhere.
I think the axis of AI in Ed should be, use it to super-charge learning, while assessment should be impossible to outsource to AI.
Verbal exams.
ghaff
Assessments are arguably the least important part related to post-secondary education.
jespinel
> This is a bunch of fluff.
Accurate description.
mag7269
“Boy, these guys are just [your typical Bill Lumberghs] with upper management written all over them.”
jsrozner
What human writes, "This is not a moment for patches and duct tape."
Is that AI or just horribly clichéd writing?
breezybottom
Maybe JK Rowling?
copperx
I hope AI was used to generate those empty words. What a waste of human effort.
breezybottom
A bunch of highly paid administrators were hired to use AI to generate it.
tyrabound
Human slopality
breezybottom
I'm shocked, shocked, that they recommend hiring more administrators and forming more committees to study the problem.
phasefactor
"Be bold" by saying nothing of any interest. I have to agree that this is basically just fluff.
Anyone put it into pangram yet?
16bitvoid
97% human-written
phasefactor
I guess prolific em-dashes is not considered a sign of LLM writing anymore.
ghaff
I am not an LLM and use em-dashes and they were house style in my last 20+ years of employment. The whole em dashes == AI thing needs to just stop.
peri-cl
Speaking about style choices, I was curious why you sometimes use the 'f' from the Latin Extended-B block, U+0192, in your comments. I've never seen that before! (I'd meant to ignore it but you seem to be soliciting typography discussions...)
ghaff
Probably a fat finger of some sort or it came through a copy and paste. Nothing I do deliberately.
phasefactor
It is more about the document containing more than sixty of them. They are a completely legitimate piece of punctuation, but even prolific pre-LLM users did not tend to go that wild with them.
ghaff
I did not notice anything about the document in that regard that was worthy of note.
tyrabound
It's actually not em-dashes, they are en-dashes, which is either intentionally to deceive (e.g., swapped) or used incorrectly. Both of which are bad, it's up to everyone to decide on which is worse coming from MITality.
jacobolus
Spaced en dashes – like this – are a correct and accepted typographical choice. Whether to use spaced en dashes or (spaced or unspaced) em dashes is a matter of taste or local convention, and varies from one publication to another. Other than a typo or two where the space was left out on one side of a dash, the dashes in this linked document are just fine. See https://en.wikipedia.org/wiki/Dash#En_dash_versus_em_dash
FabHK
Sad that people now consider basic knowledge of typography, punctuation, and orthography a sign of AI.
phasefactor
I was not claiming they are not correct punctuation/syntax. My claim is just no one uses them that frequently... When you see dozens of them in the first few pages of a document it raises suspicion.
jaggederest
This is likely an artifact of writing standards - not especially suspicious, it's like MIT Tech Review or the New Yorker using diaresis when they need to say reëducate.
jacobolus
Some authors use a ton of dashes, while others never use dashes.
DonsDiscountGas
Do you think LLMs invented the em dash? Pre LLM, em dashes were a sign of pretentiousness. Which is why LLMs learned to use them, and why a document produced by a university has plenty of them.
ygouzerh
"Since AI excels at tasks like coding and some types of design, instructors can now assign projects that are much more ambitious." --> it's quite true.
For example, we never had time to learn well for example CI/CD in my years.
It's something that takes time to build (feedback loop is around 10 to 20 min for pipelines, instead of 1 min when coding something), and become quite costly for trial and error in term of time.
But it's now 60% of my work as DevOps Engineer, and a big part of other devs/lead devs work. They are a lot of good and bad practices.
We could imagine a class to learn about how to build proper CI/CD pipelines and experiment --> philosophies, different patterns, testing different caching strategy, etc
beej71
I do like that they're open to reconsidering the grading paradigm wholesale. I taught at a boot camp for a while and our highest-placed cohorts were the ones to whom we gave neither grades nor certificates. They plainly understood the game was between them and an eventual interviewer, not them and the school or instructor.
ghaff
There's no magic rule that you have to give grades. A few colleges don't and MIT doesn't for the (I think) first term.
zerolayers
Was expecting something more concrete with a set of deliverables that can actually be acted on.
reagle
- 2. Guiding Principles
- 2.1. Be humble
- 2.2. Be bold
- 2.3. Put humanity front and center
- 2.4. Lean into learning
- 2.5. Teach with intentionality
- 2.6. No one size fits all
- 2.7. Augmentation not automation
- 2.8. Think beyond the classroom and the campus
- 3. Recommendations
- 3.1. Adapt educational processes for an AI-aware world
- 3.1.1. Revisit course goals
- 3.1.2. Ensure durable learning through new course policies, structures, and forms of assessment
- 3.1.3. Emphasize experiential and project-based learning
- 3.1.4. Build structured in-person social learning into subjects
- 3.1.5. Preserve and expand out-of-class research and career experiences
- 3.1.6. Reconsider grades and incentives
- 3.1.7. Expand in-person spaces for labs and in-person evaluation
- 3.1.8. Provide AI use policies, with justification
- 3.1.9. Exercise caution with AI detectors and online exam platforms
- 3.1.10. Support responsible experimentation in the curriculum
- 3.2. Center people, community, and the residential experience
- 3.2.1. Define and communicate the value of residential education
- 3.2.2. Strengthen social connection and personal wellbeing
- 3.2.3. Encourage instructor disclosure around their own AI use
- 3.2.4. Teach effective, responsible, and ethical use of AI
- 3.2.5. Recognize and mitigate negative impacts of AI
- 3.2.6. Acknowledge AI use in theses and other research work
- 3.3. Build processes, teams, and tools for continuous reflection, iteration, and improvement
- 3.3.1. Establish an ongoing AI and education committee
- 3.3.2. Create school/college- or department-level AI Leads
- 3.3.3. Fund AI Fellows and an AI Implementation Team
- 3.3.4. Create an AI Pilot Fund
- 3.3.5. Provide ongoing training and instructor support
- 3.3.6. Develop metrics
- 3.3.7. Ensure equitable technology access
- 3.3.8. Protect sensitive data and preserve model choice
- 3.3.9. Establish privacy, logging, and auditing policies
- 3.3.10. Monitor AI costs and environmental impact
- 4. Conclusion
jplusequalt
I'm sick and tired of hearing about how LLMs are somehow going to cultivate a society of "creative people full of ideas" when increasingly the people who are using it the most are accepting the LLMs ideas without question.
add-sub-mul-div
It's like assuming that the advent of fast food will create a society of people who exercise and otherwise eat well, to balance it.
DarmokTanagra
Reminds me of when the crypto people were saying system used almost exclusively for money laundering was going to facilitate a financial revolution.
imagetic
vibewashing
a2ff6eeb0
Education is so cooked. In 10 years, people arent going to find it useful financially. The AI will be better at economically relevant thinking.
And there aren't enough hobbyist learners to sustain education at its current level.
beej71
I figure once it collapses I'll transition into being a tutor for some billionaire's kids at roughly triple my current salary. :)
DataDive
Isn't that how Parasite starts?
tyrabound
To be fair, the US education system has long been "cooked" due to policies implemented many decades ago now to sacrifice excellence, exceptionalism, and true merit for engineered social stability and ruling class profiteering and plunder.
AI is an opportunity to improve true education and learning on an individual and community level based on free association without the authoritarian and violent hand of government imposing lowest common denominator uniformity on everyone to produce uniform and manageable, profitable pseudo-citizen cogs.
If researchers at a well funded institution like MIT are seriously considering replacing hiring undergrads with LLMs, I can just imagine how researchers at schools with less funding are open to it. I never did research as part of my undergrad (something I regret, but I likely wasn’t in the headspace for it back then) but my friends who did view it as a core part of their education and deeply helpful for future opportunities. I really hope schools come up with a policy to help discourage this.
But undergrad research has always served two roles: getting my work done and the pleasure of giving the best students the chance to go beyond the curriculum, which is designed for the average student.
I think the impact of AI on undergrad research opportunities will be zero. The bigger impact is reduced science funding. I have directed lots of PhD student grant funding to undergrads who deserved it. I don’t see myself doing that in the near future.
"My leg already had a cut on it. No need to worry about the new stab wound in my chest!"
A document like this isn't going to magically "solve" the use of AI in higher education. The purpose is to define a shared understanding of the situation across a large, complex organization, and set an initial direction and general shape for actions to take.
It contains clear, specific observations of how AI is organically changing the reality of education. And, in my opinion, fairly clear high-level guidance on what MIT as an entity wants to do about AI, and what individual departments and faculty should decide on their own.
A document like this doesn't need to be revolutionary, it may feel like fluff because no specific item in it is particularly surprising or groundbreaking, but the value is in having the entire document as a whole. And it is a lot more comprehensive and well thought out than what most companies can put out.
This is dozens of pages long and reading it thoroughly will net you maybe a dozen useful sentences. That ratio is the problem people are complaining about.
No, I can't provide a cite. Insider information. But whistleblowers have been threatened with being "Shwartzed" since (e.g. have frivolous criminal charges filed, to make their lives hell). In every case I've seen, whistleblowers signed an NDA, a non-disparage, and backed down.
That's why I don't trust a !@#$% thing from MIT anymore. Lies and fraud get covered up. Once that starts happening, you can't trust anything.
In this particular case that makes the message more trustworthy for me ;-)
Tell us more about how MIT giving AI some thought compares superiorly to 'many German universities', and please tell us more about which German universities and in which circumstances.
I'm just ignorant to German universities. Honestly, the only thing I know of is not even University level German education, but just the concept of the German gymnasium system, and that was from when a exchange student visited my high school, in the 1990s!
Thanks for the response. Cheers.
To explain briefly:
Broadly speaking, there are these types of schools:
- Hauptschule (lower secondary school) - Realschule (intermediate secondary school) - Gymnasium (grammar school)
in various organizational forms. At a Gymnasium, you can obtain a full Abitur (university entrance qualification), which qualifies you to study at - Universities - Universities of Applied Sciences
For studies at Universities of Applied Sciences, a Fachabitur (subject-specific university entrance qualification) is necessary, which can be obtained in various ways. Previously, diplomas were usually awarded there, an academic degree between a B.A. and M.A. Since the Bologna reforms, the system has been changed to B.A./M.A.
At Universities of Applied Sciences, studies are more practically oriented (with practical phases); however, the biggest difference is that you cannot pursue a doctorate there, which is only possible at a "proper" university.
Besides social informatics, I also teach academic writing, which is why the topic of AI is important to me, because it's not easy for teachers to deal with. But I also don't think the strategy of simply looking away is effective.
"Guiding principles: Be bold. Be humble. Put humanity front and center. Lean into learning. Teach with intentionality. No one size fits all."
"Recommendations: Adapt educational processes for an AI-aware world. Center people, community, and the residential experience. Build processes, teams and tools for continuous reflection, iteration, and improvement"
Yes, it's 95% fluff, but there is a real admission here that the guidance really ought to accept that there will be a whole host of tasks/assignments that kids engage in where the assumption SHOULD be that they basically will leverage AI, and that for their own benefit, resources should be set aside to promote experiences -- both in the context of assessments but also even in the social sphere -- there the influence of AI will be very purposefully prevented.
That is an actual stake in the ground, I think (as far as it goes in context like this). I think I would have been hard-pressed to have predicted that there would be such a revolutionary technology where the explicit ask of MIT faculty would be to prevent its influence in the school.
It seems as if all of teaching and student evaluation needs to be rebuilt from the ground up with AI as a default assumption.
Local high schools have added in class timed, hand written, essays for classes like AP US History.
I particularly like Chicago Law School's policy on AI.
E.g. Law students writing a "Substantial Research Paper" will have to defend it orally: "We will be adding one additional requirement, which is that all students will be required to engage in an oral discussion of their SRP with their supervising professor, in an in-person setting. This discussion will occur after a complete draft (or final version) of the paper has been submitted to the professor. The discussion could take place one-on-one, or as a class presentation in the style of an academic workshop. Either way, the oral exchange will involve the student answering questions that probe the reasoning of the paper and the implications of its arguments."
https://www.law.uchicago.edu/news/ai-strategy-statement
For education I can’t imagine. You might have to resort to actual discussion in classrooms instead of homework, with the accompanying need for both more educators and higher quality therein. Which isn’t going to be easy when education is fully under regulatory capture (at least for K-12 in the USA)
Which works in some contexts; I've taken seminars.
Not sure how that works with a multi-hundred person core calculus, physics, or chemistry class.
K-12 means the public school system in USA & canada. It consists of kindergarten, elementary , middle and high schools, encompassing 12 grades/classes, each taking one year to complete.
If you basically assume that everyone can and will cheat at all times, then you can let yourself be far more agnostic about it AS LONG as you are confident in some ability to measure the person's skill without the tools.
I think we need to move past recall as the test itself, and tasks as the gate we judge academic achievement by. These were only ever an approximate measure, which is why we still have things like thesis defenses.
But imagine the final being an interview with an agent that is designed to probe your mastery? From my experience so far it seems we could measure much more directly the educational outcome and also accelerate the learning at the same time. But this assumes those involved want to learn, which widespread cheating suggests is not the case. Maybe the real answer is that we shouldn’t attempt to force curiosity but instead cultivate it where we find it, allowing those who reject it to be redirected until they find something engaging.
> Students should not feel policed. Durable change will require instructors to be as clear as possible about their expectations and students to understand AI misuse as an unacceptable deviation from shared peer norms and community values rather than a violation of an arbitrary bureaucratic rule.
https://www.philanthropy.com/news/can-a-350-million-gift-cha...
All concerns, while the real goal is stated right in the article:
"With this new approach, he says, experts in many fields can gain a deeper understanding of AI so they can better harness it. Meanwhile, computing experts are gaining greater exposure to the work of their counterparts in other fields. That exposure is giving the technologists a better understanding of how to create and train AI tools to better serve others."
I'm sure they will have many meetings about future meetings.
MIT is so very, very overcapitalized.
These things don't lead to anything other than a rich guy getting his name on a building, and more corruption in academia. By far the best thing which could happen to MIT would be for 95% of the endowment to go up in a poof of smoke, bringing it back to nineties levels. That's the end of the period when the Institute did high-integrity research.
I want to get back to the era before people went to MIT as a center of wealth, power, and prestige, and would lie and cheat to get tenured positions there, to where it was a place where nerds went to nerd together.
And to a place where every !@#$% done didn't lead to a self-hyping press release, but where statements were precise, measured, calibrated, and most importantly, scientifically-accurate.
1. Upvotes
2. Time passes
3. Downvotes
I'm starting to suspect Institutional astroturf.
That fact, multiplied over 1000 different disciplines, is one of the ways that AI is going to bring amazing progress to our world.
I think the axis of AI in Ed should be, use it to super-charge learning, while assessment should be impossible to outsource to AI.
Verbal exams.
Accurate description.
Is that AI or just horribly clichéd writing?
Anyone put it into pangram yet?
For example, we never had time to learn well for example CI/CD in my years.
It's something that takes time to build (feedback loop is around 10 to 20 min for pipelines, instead of 1 min when coding something), and become quite costly for trial and error in term of time.
But it's now 60% of my work as DevOps Engineer, and a big part of other devs/lead devs work. They are a lot of good and bad practices.
We could imagine a class to learn about how to build proper CI/CD pipelines and experiment --> philosophies, different patterns, testing different caching strategy, etc
- 2.1. Be humble
- 2.2. Be bold
- 2.3. Put humanity front and center
- 2.4. Lean into learning
- 2.5. Teach with intentionality
- 2.6. No one size fits all
- 2.7. Augmentation not automation
- 2.8. Think beyond the classroom and the campus
- 3. Recommendations
- 3.1. Adapt educational processes for an AI-aware world
- 3.1.1. Revisit course goals
- 3.1.2. Ensure durable learning through new course policies, structures, and forms of assessment
- 3.1.3. Emphasize experiential and project-based learning
- 3.1.4. Build structured in-person social learning into subjects
- 3.1.5. Preserve and expand out-of-class research and career experiences
- 3.1.6. Reconsider grades and incentives
- 3.1.7. Expand in-person spaces for labs and in-person evaluation
- 3.1.8. Provide AI use policies, with justification
- 3.1.9. Exercise caution with AI detectors and online exam platforms
- 3.1.10. Support responsible experimentation in the curriculum
- 3.2. Center people, community, and the residential experience
- 3.2.1. Define and communicate the value of residential education
- 3.2.2. Strengthen social connection and personal wellbeing
- 3.2.3. Encourage instructor disclosure around their own AI use
- 3.2.4. Teach effective, responsible, and ethical use of AI
- 3.2.5. Recognize and mitigate negative impacts of AI
- 3.2.6. Acknowledge AI use in theses and other research work
- 3.3. Build processes, teams, and tools for continuous reflection, iteration, and improvement
- 3.3.1. Establish an ongoing AI and education committee
- 3.3.2. Create school/college- or department-level AI Leads
- 3.3.3. Fund AI Fellows and an AI Implementation Team
- 3.3.4. Create an AI Pilot Fund
- 3.3.5. Provide ongoing training and instructor support
- 3.3.6. Develop metrics
- 3.3.7. Ensure equitable technology access
- 3.3.8. Protect sensitive data and preserve model choice
- 3.3.9. Establish privacy, logging, and auditing policies
- 3.3.10. Monitor AI costs and environmental impact
- 4. Conclusion
And there aren't enough hobbyist learners to sustain education at its current level.
AI is an opportunity to improve true education and learning on an individual and community level based on free association without the authoritarian and violent hand of government imposing lowest common denominator uniformity on everyone to produce uniform and manageable, profitable pseudo-citizen cogs.