Prism

The Post-LLM World: Fighting Digital Garbage https://archive.org/details/paper_20260127/mode/2up

Mini paper: that future isn’t the AI replacing humans. its about humans drowning in cheap artifacts. New unit of measurement proposed: verification debt. Also introduces: Recursive Garbage → model collapse

a little joke on Prism)

I run a small print-on-demand platform and this is exactly what we're seeing. The submissions used to be easy to filter with basic heuristics or cheap classifiers, but now the grammar and structure are technically perfect. The problem is that running a stronger model to detect the semantic drift or hallucinations costs more than the potential margin on the book. We're pretty much back to manual review which destroys the unit economics.

What they mean by "academic" is fairly limited here, if LaTeX is the main writing platform. What are their plans for expanding past that, and working with, say Jane Biomedical Researcher with a GSuite or Microsoft org, that has to use Word/Docs and a redlining-based collaboration workflow? I can certainly see why they're making it free at this point.

FWIW, Google Scholar has a fairly compelling natural-language search tool, too.

Not gonna lie, I cringed when it asked to insert citations.

Like, what's the point?

You cite stuff because you literally talk about it in the paper. The expectation is that you read that and that it has influenced your work.

As someone who's been a researcher in the past, with 3 papers published in high impact journals (in chemistry), I'm beyond appalled.

Let me explain how scientific publishing works to people out of the loop:

- science is an insanely huge domain. Basically as soon as you drift in any topic the number of reviewers with the capability to understand what you're talking about drops quickly to near zero. Want to speak about properties of helicoidal peptides in the context of electricity transmission? Small club. Want to talk about some advanced math involving fourier transforms in the context of ml? Bigger, but still small club. When I mean small, I mean less than a dozen people on the planet likely less with the expertise to properly judge. It doesn't matter what the topic is, at elite level required to really understand what's going on and catch errors or bs, it's very small clubs.

2. The people in those small clubs are already stretched thin. Virtually all of them run labs so they are already bogged down following their own research, fundraising, and coping with teaching duties (which they generally despise, very few good scientist are barely more than mediocre professors and have already huge backlogs).

3. With AI this is a disaster. If having to review slop for your bs internal tool at your software job was already bad, imagine having to review slop in highly technical scientific papers.

4. The good? People pushing slop, due to these clubs being relatively small, will quickly find their academic opportunities even more limited. So the incentives for proper work are hopefully there. But if asian researchers (yes, no offense), were already spamming half the world papers with cheated slop (non reproducible experiments) in the desperate bid of publishing before, I can't imagine now.

To be fair, the question “what will change” does not presume the changes will be positive. I think it’s the right question to ask, because change is coming whether we like it or not. While we do have agency, there are large forces at play which impact how certain things will play out.