Eric EvansMar 10
Lawprofblawg

I used to waste years preparing gourmet meals. It was creative, an expression of my humanity, and fostered zen.

Thanks to McDonalds, I don't have to do that!

#Satire

Mar 10 at 4:03amWeb
Show threadM.S. Bellows, Jr.Mar 10
@lawprofblawg We used to waste time having sexual intercourse. Now my AI can track my wife's cycle and calculate when to trigger the automatic artificial inseminator.
Show threadMastodon MigrationMar 10

@lawprofblawg

This is sad

Show threadIm Having A Good Day (assertive)Mar 10
@lawprofblawg you know how every now and then there'll be a paper or an article like "yeah so it turns out the entire medical community believed this inaccurate thing because of this one paper from the 70s and no one ever questioned it"? lmao
Show thread💀 𝓕airchild 💀Mar 10
@lawprofblawg Awesome. So at the end of this scenario this guy would not have learned shit. His PhD cake has been baked by a machine, but he could present it as his own.
Show threadSam VinesMar 10
@tankgrrl @lawprofblawg imagine trying to defend your PhD thesis when Claude wrote most of it... (usually a 2-3hr interview where they ask you probing questions about your thesis to make sure you understand what you wrote and that you did in fact write it...)

"Urm... Yes I did it that way because... 🤤🤤🤤"

What an absolute idiot
Show threadTyrone SlothropMar 10

@lawprofblawg It used to be that you had to spend thousands of dollars on paying someone in the Global South to write your PhD thesis for you.

Now all you need is a subscription to Claude or OpenAI!

Show threadTyrone SlothropMar 10

@lawprofblawg Hypothesis:

Anyone who genuinely believes that AI can “replace” humans is living such a limited existence that this is indeed possible.

Show threadrandom_regretMar 10
@slothrop @lawprofblawg YESS I've been saying that for years. The only reason why it seems plausible to people that AI would replace artists, writers and academics is because our collective experience of art, reading and science has been degraded for decades by capitalist market forces. The question is how could we have let it come so far that our cultural output and, even more importantly, our cultural literacy has become so overwhelmingly underwhelming?
Show threadJWcph, Radicalized By DecencyMar 10
@lawprofblawg Way worse. "I spent years learning how to prepare gourmet food but this machine turns ingredients directly into shit, no eating or digestion necessary - oh welp"...
Show threadCassandrichMar 10

@lawprofblawg It's not even McDonalds. It's like pieces of foam painted to look like food. It doesn't even perform the function it's purported to on a minimal level. It just looks like the same thing to a really noncritical observer.

That comment was written by someone who utterly fails to understand what research *is*.

Show threadkrigMar 10

@lawprofblawg I used to spend years lifting heavy weights, slowly building my strength and endurance, fighting through fatigue and lack of motivation.

A forklift could have lifted all that weight in an instant.

The weight lifting game has changed forever.

Show threadTDayMar 10
@krig @lawprofblawg Ah a perfect analogy .
Show threadPedestrianError  Mar 10

@krig @lawprofblawg I used to spend 10 minutes walking to the neighborhood grocery store. On the way I met neighbors, cleaned up litter, saw interesting flowers and artwork people on people's homes and businesses, got exercise, and didn't pollute the air, make loud noise, put anyone's life at risk.

A car could get me there in 2 minutes. The transportation game has changed forever. Are we really proud of that in hindsight?

Show threadJustin Macleod2d ago
@krig @lawprofblawg It's almost as if this post isn't about weight-lifting at all and has some deeper meaning. Wonder what it could be! I know, I'll ask Chat GPT. 💡
Show threadDyrathror ⁂ 🇺🇦Mar 10
@lawprofblawg
He completely missed the point what studies are for.
I hope he returns his PhD and resigns from his post.
Show threadWellsiteGeoMar 10
@lawprofblawg And you don't have to worry about food poisoning for your guests - all disease can be blamed on McDonalds. Which makes it good.
Show threadbovazMar 10
@lawprofblawg but if the llm had read for him, what would he know?
Also, what kind of experiment would an llm be able to perform and test? None in the real world that I can think of.
(All that, on top of all the other issues with llms)
Show threadJulian BenderMar 10

@bovaz @lawprofblawg They are trying:
https://www.biorxiv.org/content/10.1101/2025.08.01.668204v1.full.pdf+html

It’s great to automate this, the reason we need LLMs for multidimensional optimisation when we have several numerical approaches for this is still unclear to me though.

Design-driven optimization of low-cost reagent formulations for reproducible and high-yielding cell-free gene expression

Access to recombinant proteins is vital in basic science and biotechnology research. Cell-free gene expression systems provide one approach to address this need, but widespread utilization remains limited by the cost, complexity, and inconsistency of current platforms. To address these limitations, we carry out a multi-dimensional definitive screening design to reduce the number of reagent components and remove costly secondary energy substrates. From more than 1,200 reagent formulations, we discover a simple and reproducible system based on 12 components. The optimized reagent formulation can produce 2.4 ± 0.3 g/L of protein product at the 15-μL scale (∼$55/gprotein) and 3.7 ± 0.2 g/L (∼$36/gprotein) at the 4-mL scale with oxygen supplementation. This provides an 84 to 99% reduction in cost over previous cell-free reagent formulations. We further show that the optimized reagent formulation can produce nucleoside triphosphates from nucleotides and ribose and that it is robust to failure across batches of cell lysates, users/locations, and in the synthesis of different proteins. Specifically, we demonstrate the production of fifteen therapeutically relevant products, including full-length aglycosylated monoclonal antibodies. We anticipate that our optimized reagent formulation will further democratize the use of cell-free systems for protein manufacturing and synthetic biology applications. ### Competing Interest Statement MLO and MCJ have filed an invention disclosure based on the work presented. MCJ has a financial interest in SwiftScale Biologics, Gauntlet Bio, Pearl Bio, Inc., Synolo Therapeutics and Stemloop Inc. MCJ's interests are reviewed and managed by Stanford University in accordance with their competing interest policies. National Science Foundation, CBET - 2341123, DGE-1842165, 2021900 DARPA, W911NF-23-2-0039 Department of Energy, DE-SC0023278

bioRxiv
Show threadbovazMar 11
@aretaon @lawprofblawg probably the same reason why we "need" llms for about anything else they sell them for.
Show threadChrisMar 11
@bovaz @lawprofblawg No need, it just bullshits about the experiment *and* the result.
Show threadbovazMar 11
@thechris @lawprofblawg but any competent student can already do that...
Show threadChrisMar 11
@bovaz @lawprofblawg Well, of course. But the bullshit machine lets you cheat with much less effort.
Show threadHans-HelgeMar 10
@lawprofblawg Wait to the moment Prof Lennart Nacke realizes, that ChatGPT can "read" these papers, but that does not mean, he as a scientist gains the knowledge.
Show threadRetrogradeMar 10
@lawprofblawg The research game? 
Show threadStumpy The MuttMar 10
@lawprofblawg I got my PhD free in a box of cereal. All I have to do now is ask Claude questions, but I'm too lazy to be bothered doing even that.
Show threadSeñor Baby, Sr.Mar 10
@lawprofblawg hahaha, of course it's Nacke...
Show threadTrump & Epstein: Biz partners.Mar 10

@lawprofblawg
There is a tremendous danger here that people start relying on #AI so much they don't bother to learn on their own, and eventually unable to "fact check" what they are told to know if it's correct or not. 🤔

"#ArtificialIntelligence will never be a match for Natural Stupidity" - Mugsy

Show threadyakMar 10
@lawprofblawg I'm 100% confident that if I had gotten chatgpt to write my dissertation it would be full of subtly incorrect arguments and typos.

I mean, I wrote tons of subtly incorrect arguments and typos, but at least I knew every step in every proof and I could check it easily and go back and fix things relatively easily.
Show threadDavid ShenemanMar 10
@lawprofblawg @aleen why do the work when you can skip the work and get a prize as if you had? 🤦🏼‍♂️
Show threadDr Andrew A. Adams #FBPE 🔶Mar 11
@lawprofblawg @publicvoit
Most primary school students waste weeks doing repetitive arithmetic calculations that a pocket calculator or calculator app could crush in seconds…
The process is the point, not just the immediate product.
Show threadGuillotine Jones, FlâneurMar 11
@lawprofblawg
Thanks to Wonder Bread, I don't have to bother with sourdough starter anymore..
Show threadMegan Lynch (she/her)Mar 11
@lawprofblawg ...
Show threadMartin Ueding6d ago

@lawprofblawg
I've done a PhD in computational physics prior to AI, now I work as a staff software engineer and conservatively use AI coding tools.

The person in the screenshot has a point. In my PhD, there were many time consuming and stupid tasks. Like figuring out dependency paths to compile a library. Porting plotting code from one library to another. This tedium is not where research happens.

If you delegate your core job, it will be bad, of course.

Show threadtkissing6d ago
@lawprofblawg Guy got his degree in Bielefeld. I always thought the Bielefeld Conspiracy was made up, but maybe it's true after all. I don't *want* a real German school to give a PhD to someone with such a broken understanding of what research is.
Show threadEl Duvelle2d ago

@lawprofblawg
The quoted post is also completely wrong - a gen AI program cannot design, run and report experiments. it can certainly write up some text that says that it did...
Of course, it would be nice to accelerate some parts of research and we already do that via automation of some parts of experiments and of data analysis. But inventing the contents of research is the opposite of what we want.

So we definitely need to keep training researchers, not only because doing it yourself is nice and formative, but also because it is the only way to actually do it.

Show threadFalow, Master Surchid2d ago

@lawprofblawg

I spent years trying to learn the French language, to understand its history and culture, so that I could travel to and appreciate French-speaking lands.

Now, the translate app on my phone can let me understand what they are saying in seconds.

Learning a foreign language has changed forever.

Show threadbipolaron1d ago
@lawprofblawg nauseating. I guess it's job security for me, someone whose job is largely helping fix people's broken experiments.