Prism

Have you ever seen the comment section of a Snowden thread here? A lot of users here call for Snowden to be jailed, call him a russian asset, play down the reports etc. These are either NSA sock puppet accounts or they won't bite the hand that feeds them (employees of companies willing to breach their users trust).

Edit: see my comment here in a snowden thread: https://news.ycombinator.com/item?id=46237098

We haven’t forgotten… it’s mostly that we’re all jaded given the fact that there has been zero ramifications and so what’s the use of complaining - you’re better off pushing shit up a hill

Yes, imho, there is a great deal of ignorance of the actual contents of the NSA leaks.

The agitprop against Snowden as a "Russian agent" has successfully occluded the actual scandal, which is that the NSA has built a totalitarian-authoritarian apparatus that is still in wide use.

Autocrats' general hubris about their own superiority has been weaponized against them. Instead of actually addressing the issue with America's repressive military industrial complex, they kill the messenger.

(full disclosure, yes they will be handin in PII on demands like the same kinda deals, this is 'normal' - 2012 shows us no one gives a shit)

I just think it's silly to obsess over words like that. There are many words that take on different meanings in different contexts and can be associated with different events, ideas, products, time periods, etc. Would you feel better if they named it "Polyhedron"?

I personally associate Prism with [Silverlight - Composite Web Apps With Prism](https://learn.microsoft.com/en-us/archive/msdn-magazine/2009...) due to personal reasons I don't want to talk about ;))

[1]: https://openai.com/index/openai-appoints-retired-us-army-gen...

There's a good chance they just asked GPT5.2 for a name. I know for a fact that when some of the OpenAI models get stuck in the "weird" state associated with LLM psychosis, three of the things they really like talking about are spirals, fractals, and prisms. Presumably, there's some general bias toward those concepts in the weights.

If it was part of their adtech systems and them dipping their toe into the enshittification pool, it would have been a legendarily tone deaf project name, but as it is, I think it's fine.

It's a horrible name for any product coming out of a company like OpenAI. People are super sensitive to privacy and government snooping and OpenAI is a ripe target for that sort of thinking. It's a pretty bad association. You do not want your AI company to be in any way associated with government surveillance programs no matter how old they are.

I do self-host Overleaf which is annoying but ultimately doable if you don't want to pay the $21/mo (!).

I do have to wonder for how long it will be free or even supported, though. On the one hand, remote LaTeX compiling gets expensive at scale. On the other hand, it's only a fraction of a drop in the bucket compared to OpenAI's total compute needs. But I'm hesitant to use it because I'm not convinced it'll still be around in a couple of years.

[0]: https://typst.app

And you think the indians will not hand write the output of LLMs ?

Not that I have a better suggestion myself..

Exactly, and I think this is good news. Let's break it so we can fix at last. Nothing will happen until a real crisis emerges.

I actually think Prism promotes a much more responsible approach to AI writing than "copying from chatgpt" or the likes.

"Human Verification as a Service": finally, a lucrative career where the job description is literally "read garbage all day and decide if it's authentic garbage or synthetic garbage." LinkedIn influencers will pivot to calling themselves "Organic Intelligence Validators" and charge $500/hr to squint at emails and go "yeah, a human definitely wrote this passive-aggressive Slack message."

The irony writes itself: we built machines to free us from tedious work, and now our job is being the tedious work for the machines. Full circle. Poetic even. Future historians (assuming they're still human and not just Claude with a monocle) will mark this as the moment we achieved peak civilization: where the most valuable human skill became "can confidently say whether another human was involved."

Bullish on verification miners. Bearish on whatever remains of our collective attention span.

This appears to just be the output of LLMs itself? It credits GPT-5.2 and Gemini 3 exclusively as authors, has a public domain license (appropriate for AI output) and is only several paragraphs in length.

Maybe you get reimbursed for half as long as there are no obvious hallucinations.

https://hn.algolia.com/?dateRange=pastYear&page=0&prefix=tru...

https://hn.algolia.com/?dateRange=pastYear&page=0&prefix=tru...

This is a space that probably needs substantial reform, much like grad school models in general (IMO).

On the other hand, the world is now a different place as compared to when several prominent journals were founded (1869-1880 for Nature, Science, Elsevier). The tacit assumptions upon which they were founded might no longer hold in the future. The world is going to continue to change, and the publication process as it stands might need to adapt for it to be sustainable.

For developers, academics, editors, etc... in any review driven system the scarcity is around good human judgement not text volume. Ai doesn't remove that constraint and arguably puts more of a spotlight on the ability to separate the shit from the quality.

Unless review itself becomes cheaper or better, this just shifts work further downstream and disguising the change as "efficiency"

> > who are looking to 'boost' their CV

Ultimately, this seems like a key root cause - misaligned incentives across a multi-party ecosystem. And as always, incentives tend to be deeply embedded and highly resistant to change.

These acts just must have consequences so people stop doing them. You can use AI if you are doing it well but if you are wasting everyones time you should just be excluded from the discourse altogether.

This is still a good step in a direction of AI assisted research, but as you said, for the moment it creates as many problems as it solves.

I'm not sure what the final state would be here but it seems we are going to find it increasingly difficult to find any real factual information on the internet going forward. Particularly as AI starts ingesting it's own generated fake content.

So I was not amused about this announcement at all, however easy it may make my own life as an author (I'm pretty happy to do my own literature search, thank you very much).

Also remember, we have no guarantee that these tools will still exist tomorrow, all these AI companies are constantly pivoting and throwing a lot of things at the wall to see what sticks.

OpenAI chose not to build a serious product, as there is no integration with the ACM DL, the IEEE DL, SpringerNatureLink, the ACL Anthology, Wiley, Cambridge/Oxford/Harvard University Press etc. - only papers that are not peer reviewed (arXiv.org) are available/have been integrated. Expect a flood of BS your way.

When my student submit a piece of writing, I can ask them to orally defend their opus maximum (more and more often, ChatGPT's...); I can't do the same with anonymous authors.

"which is really not the point of these journals at all"- it seems that it very much is one of the main points? Why do you think people publish in journals instead of just putting their work on the arxiv? Do you think postdocs and APs are suffering through depression and stressing out about their publications because they're agonizing over whether their research has genuinely contributed substantively to the academic literature? Are academic employers poring over the publishing record of their researchers and obsessing over how well they publish in top journals in an altruistic effort to ensure that the research of their employees has made the world a better place?

Plenty of researchers hate writing and will only do it at gunpoint. Or rather, delegate it all to their underlings.

I don't see an issue with generative writing in principle. The Devil is in the details, but I don't see this as much different from "hey grad student, write me this paper". And generative writing already exists as copy-paste, which makes up like 90% of any random paper given the incrementality of it all.

I was initially a little indignated by the "find me some plausible refs and stick them in the paper" section of the video but, then again, isn't this what most people already do? Just copy-paste the background refs from the colleague's last paper introduction and maybe add one from a talk they saw in the meantime, plus whatever the group & friends produced since then.

My experience is most likely skewed (as all are), but I haven't met a permanent researcher that wrote their own papers yet, and most grad students and postdocs hate writing. Literally the only times I saw someone motivated to write papers (in a masochistic way) were just before applying to a permanent position or while wrapping up their PhD.

Onto your point, though, I agree this is somewhat worrisome in that, by reaction, the barrier to entry might rise by way of discriminating based on credentials.

For whom? For OpenAI these tools are definitely the solutions. They are developing by throwing various AI-powered stuff at the wall to see what sticks. These tools also demonstrate to the investors that innovation did not stall and to show that AI usage is growing.

Same with Microsoft: none of the AI stuff they are shoving down the users' throats were actually designed for the users. All this stuff is only for the token usage to grow for the shareholders to see.

Similar with Google although no one can deny real innovation happening there.

It was already a problem 25 years ago when I did my Ph.D., and I don't think things changed that much since then.

This encourages researchers to publish barely valuable results, or to cut one articles into multiple ones with small variations to increase their number of publications. Also publishers creating more conferences and more journals to respond to the need that researchers have to publish.

I remember many experienced professors telling me cynically about this, about all the techniques they had to blow up one small finding into many articles.

Anyway - research slop started way before AI. It's probably going to make the problem worse, but the root issue have been there for a long time.

The target audience of this tool is not academics; it's OpenAI investors.

So yes, you use it to write the paper but soon it is public knowledge anyway.

I am not sure if there is much to learn from the draft of the authors.

[1]: https://statmodeling.stat.columbia.edu/2026/01/26/machine-le...

https://www.graphpad.com/

To clarify, there is a difference between a bibliography (a list of relevant works but not necessarily cited), and cited work (a direct reference in an article to relevant work). But most people start with a bibliography (the superset of relevant work) to make their citations.

Most academics who have been doing research for a long time maintain an ongoing bibliography of work in their field. Some people do it as a giant .bib file, some use software products like Zotero, Mendeley, etc. A few absolute psychos keep track of their bibliography in MS Word references (tbh people in some fields do this because .docx is the accepted submission format for their journals, not because they are crazy).

Obviously ridiculous, since a philosophical argument should follow a chain of reasoning starting at stated axioms. Citing a paper to defend your position is just an appeal to authority (a fallacy that they teach you about in the same class).

The citation requirement allowed the class to fulfill a curricular requirement that students needed to graduate, and therefore made the class more popular.

You're right that something like Cursor can work if you're familiar with all the requisite tooling (git, installing cursor, installing latex workshop, knowing how it all works) that most researchers don't want to and really shouldn't have to figure out how to work for their specific workflows.

I think I would only switch from Overleaf if I was writing a textbook or something similarly involved.

@vicapow replied to keep the Dropbox parallel alive

I have a phd in economics. Most researchers in that field have never even heard of any of those tools. Maybe LaTeX, but few actually use it. I was one of very few people in my department using Zotero to manage my bibliography, most did that manually.

generally think that there's a lot of fertile ground for smart generalist engineers to make a ton of progress here this year + it will probably be extremely financially + personally rewarding, so I broadly want to create a dedicated pod to highlight opportunities available for people who don't traditionally think of themselves as "in science" to cross over into the "ai for hard STEM" because it turns out that 1) they need you 2) you can fill in what you don't know 3) it will be impactful/challenging/rewarding 4) we've exhausted common knowledge frontiers and benchmarks anyway so the only* people left working on civilization-impacting/change-history-forever hard problems are basically at this frontier

*conscious exaggeration sorry

I haven't tried it yet but Typst seems like a promising replacement: https://typst.app/

Typst feels more like the future: https://typst.app/

The problem is that so many journals require certain LaTeX templates so Typst often isn't an option at all. It's about network effects, and journals don't want to change their entire toolchain.

The main feature that's important is collaborative editing (like online Word or Google Docs). The second one would be a good reference manager.

And then I need an extra tool for dealing with bibliography, change history is unpredictable (and, IMO, vastly inferior to version control), and everything gets even worse if I open said Word file in LibreOffice.

LaTeX' syntax may be hard, but Word actively fights me during writing.

[1] Moving a photo in Microsoft Word - https://www.instagram.com/jessandquinn/reel/DIMkKkqODS5/

It is an old language though. LaTeX is the macro system on top of TeX, but now you can write markdown or org-mode (or orgdown) and generate LaTeX -> PDF via pandoc/org-mode. Maybe this is the level of abstraction we should be targeting. Though currently, you still need to drop into LaTeX for very specific fine-tuning.

The collect chat records for any number of users, not the least of which being NSA surveillance and analysis - highly likely given what we know from the Snowden leaks.

AIs use em dashes because competent writers have been using em dashes for a long time. I really hate the fact that we assume em dash == AI written. I've had to stop using em dashes because of it.

Maybe it's cynical, but how does the old saying go? If the service is free, you are the product.

Perhaps, the goal is to hoover up research before it goes public. Then they use it for training data. With enough training data they'll be able to rapidly identify breakthroughs and use that to pick stocks or send their agents to wrap up the IP or something.

The urge to cheat in order to get a job, promotion, approval. The urge to do stuff you are not even interested in, to look good in the resume. And to some extent I feel sorry for these people. At the end of the day you have to pay your bills.

Hmm, I follow the argument, but it's inconsistent with your assertion that there is going to be incentive for 'proper work' over time. Anecdotally, I think the median quality of papers from middle- and top-tier Chinese universities is improving (your comment about 'asian researchers' ignores that Japan, South Korea, and Taiwan have established research programs at least in biology).

Fuck A.I. and the collaborators creating it. They've sold out the human race.

LLMs are undeniably great for interactive discussion with content IF you actually are up-to-date with the historical context of a field, the current "state-of-the-art", and have, at least, a subjective opinion on the likely trajectories for future experimentation and innovation.

But, agents, at best, will just regurgitate ideas and experiments that have already been performed (by sampling from a model trained on most existing research literature), and, at worst, inundate the literature with slop that lacks relevant context, and, as a negative to LLMs, pollute future training data. As of now, I am leaning towards "worst" case.

And, just to help with the facts, your last comment is unfortunately quite inaccurate. Science is one of the best government investments. For every $1.00 dollar given to the NIH in the US, $2.56 of economic activity is estimated to be generated. Plus, science isn't merely a public venture. The large tech labs have huge R&D because the output from research can lead to exponential returns on investment.

Errr sure. Sounds easy when you write it down. I highly doubt such a thing will ever exist.

EDIT: Fixed :)

A comparison comes to mind is the n8n workflow type product they put out before. N8n takes setup. Proofreading, asking for more relevant papers, converting pictures to latex code, etc doesn't take any setup. People do this with or without this tool almost identically.

The reason? I can give you the full source for Sam Altman:

while(alive) { RaiseCapital() }

That is the full extent of Altman. :)

A good salesman could make money off of people who can do this, even if this is free they can always pull more than their weight with other efforts, and that can be in a more natually lucrative niche.

Look at how much BS flooded psychology but had pretty ideas about p values and proper use of affect vs effect. None of that mattered.

I'm not in favor of letting AI do my thinking for me. Time will tell where Prism sits.

In my lab, we have been struggling with automated image segmentation for years. 3 years ago, I started learning ML and the task is pretty standard, so there are a lot of solution.

In 3 months, I managed to get a working solution, which only took a lot of sweat annotating images first.

I think this is where tools like OpenCode really shine, because they unlock the potential for any user to generate a solution to their specific problem.

We removed the authorship of a a former co-author on a paper I'm on because his workflow was essentially this--with AI generated text--and a not-insignificant amount of straight-up plagiarism.

Didn't even open a single one of the papers to look at them! Just said that one is not relevant without even opening it.

"I know nothing but had an idea and did some work. I have no clue whether this question has been explored or settled one way or another. But here's my new paper claiming to be an incremental improvement on... whatever the previous state of understanding was. I wouldn't know, I haven't read up on it yet. Too many papers to write."

E.g. “cite that paper from John Doe on lorem ipsum, but make sure it’s the 2022 update article that I cited in one of my other recent articles, not the original article”

A couple of generations of students later, and these will be rare skills: information finding, actual thinking, and conveying complex information in writing.

At the end of the day, it's all about the incentives. Can we have a world where we incentivize finding the truth rather than just publishing and getting citations?

Perhaps, like the original PRISM programme, behind the door is a massive data harvesting operation.