katatue

At my (German) university, the solution would have been to change from a "cumulative" thesis (which requires peer-reviewed papers) to a monograph-style one. Because the general rule was that your undertaking must be novel ar the time you submit your thesis topic, not necessarily at the tome the thesis itself is submitted (years later).

Why do you treat unpredictability as neutral? If something has a potentially existential, but completely unpredictable outcome, surely one should avoid heading for it?

Like, knowing that there is a limited number of plausible doom scenarios and what these are, would make me more confident in taking the risk than settling on "it's literally the singularity, we can't know what happens beyond".

IA might be large enough to earn consideration, but generally scrapers just don't care about being good citizens. I work in the GLAM space and we offer OAI-PMH interfaces for the harvesting of our collections data - which doesn't stop companies from preferring to scrape our website for worse (less complete, less structured, less standardized) data instead.

Same application spectrum as CO2 fire extinguishers I'd imagine (which we use for expensive lab hardware that we don't want to get damaged by water, foam, ..., but which are utterly useless for anything more substantial becuase it just reignites).

Oulipo vindicated.

I don't know if this is really about the public sphere as ut exists today. I suspect that the core idea is more along of the lines that in a world where policy is made by artificial super intelligences, you will need your own ASI to advocate on your behalf if your convictions are to have any impact at all.

I like to let new models write a few lines of Latin poetry - they rarely get the meter right.

I don't have access to paid ChatGPT right now, but here's Opus 4.6 with extra thinking enabled: https://claude.ai/share/6e0e8ef5-06e4-4514-ba7e-299357c1fc55

The initial draft fucks up the meter in lines 3 and 8, the final version gets line 2 wrong ("venit meis") and is somewhat obnoxious with verses 2 and 8 basically repeating each other. The thinking trace is useless and gives us no clue why the model exchanged a bland, but metrically correct first distich for a more interesting, but metrically incorrect one.

In fact, the "careful" examination of its own output completely skips the erroneously modified half-verse in line 2 - now, tell me that's a coincidence and not a sign of bullshitting.