> Don't forget that you can adjust your requirements (either via plan or skill) to ensure the mistakes do not happen.

No, you can't. Adjusting prompts ensures absolutely nothing.

I disagree. What I should have added is that with agents (as well as humans) you do need to have tests that verify what was done.

That assumes you can write automated tests that reliably identify the mistakes over an entire codebase. Nice idea in theory. If it were actually possible, we would long since have generalized libraries of tests to catch every significant security and performance gotcha. What we have are static code analysis tools, fuzzers, etc. None of which have come close to eliminating security and performance problems. I don't see how AI somehome changes that.

Ah, I see what you mean now. Yes, my mind went straight to static analysis and testing (unit, feature, uat, mutation). Thanks for expanding on your point!

> I dunno, for decades the policy by most of the West has been (a) keep Germany from re-arming in case they start WW3

That policy lasted less than a single decade. Germany was encouraged to re-arm as soon as 1950 inofficially and 1955 officially.

And I think even before 1950 there was a feeling, particularly in the US, that it was good to have the Germans on-side in a military conflict due to their recent experience fighting the Soviets.

Someone who saw the $$$ previously spent on humans to do it.

Lovely! How was the mechanical setup to ensure that all those shots are consistent, and how long did it take?

I posted some pictures.. takes only 20 min!

Dogme is Danish, because it's a manifesto of a founded in Denmark - and it's basically an update of https://en.wikipedia.org/wiki/Dogme_95

I don't think it's meant as a constraint to be worked around, but as a guardrail against being inauthentic.

And it excludes a lot less than its inspiration Dogme 95, which has as one rule "Genre movies are not acceptable."

> "Genre movies are not acceptable."

I find that hilarious, like proclaiming that only other people have an ethnicity or an accent. Because of course Dogme is a genre of its own.

Perhaps Dogme 95/Dogma 25 films are in a genre of their own, but they're not "genre movies." People make the same argument with "literary fiction"/"non-genre fiction" vs "genre fiction." The terms have meaning whether or not you want to acknowledge it.

Dogme is more of a methodology than genre. Genre usually means settings and tropes, like scifi or horror or superhero.

Though I’d argue that rom-com, period pieces, and biopics also are “genre”, at least to the extent a particular movie just paints by numbers within those styles.

There's actually Street View images, so you can take a look, also at the agricultural plots southwest of the town (see https://en.wikipedia.org/wiki/Potato_Patches ). There's some sheep, cattle and (I think) donkeys as well.

> Seriously? You can't fathom an honest researcher asking for AI to find a citation they know exists, and the AI inserting or modifying a citation incorrectly without them realizing?

Indeed I cannot. If you do that, you are not, in fact, an honest researcher. You're a lazy hack.

> You’re right that a single hallucinated line is not evidence of reckless disregard

It absolutely is.

> - because that could have happened on a final follow-up pass after you had performed due diligence.

A "final follow-up pass" that lets the LLM make whatever changes it deems appropriate completely negates all the due diligence you did before, unless you very carefully review the diffs. And a new or substantially changed citation should stand out in that diff so much that there's no possible excuse to missing it.

> It’s happened to me.

Then you were guilty of reckless disregard.

> I know how challenging it can be to keep bad patterns out of LLM generated output

If your research paper contains any LLM generated output you did not manually vet, you are a hack and should not get published.

Context matters a lot. I didn’t publish any papers, and didn’t even provide the context of the LLM mistakes I caught to later report on, so no, I was not guilty of reckless disregard. I found those mistakes on an acceptable timeline.