你这么说好帅

”all language is approximation” believers when my autistic ass finally weaves together the exact right linguistic phrase that conveys pure information and shatters qualia as we know it

From now on this is how all petty fandom disputes should be settled.

Not enough people talking about how well broccoli holds a sauce

I wish people were as scared of getting into a car accident as they are of being true crime'd. Maybe then they wouldn't be on their phones while driving.

Let’s fucking go

5. If the words are Google's, this solidifies the position of universities who demand that all answers from AI are fully cited. If all the in-line citations now have to be (Google, 2026), that's going to make it obvious when someone's trying to use Google as a source. There's still the difficulty with people who are academically dishonest by trying to pass off the AI writing as their own. 6. 91% accuracy is officially too low to use as a source of references, which means the AI can't be used as a source of references either. This makes it less legitimate for such purposes than Wikipedia of all places (Wikipedia might need date/time proof of when it was accessed for the reference to be valid, but at least it is possible to prove the link existed at a particular date and time). 7. This will help encourage the rollout of courses on how to avoid AI search for students who need academic accuracy, because it's statistically not good enough to use. 8. This strengthens the case intellectual property authors have against Google in the EU, as this is proof that an intellectual property transfer took place.

you have to preface things like this with disclaimers that you are not pro-ai for people incapable of critical thinking but: its so awesome how much people post about AI having no soul and lacking humanity and merely mimicking some sort of innate Humanness while failing to address or contend with the fact that a lot of people are seemingly making work indistinguishable from the statistical average spit up by a computer.

ppl could ABSOLUTELY churn out bland slop that's trying very hard to scan as though it's legit way before AI showed up, and i stopped trying to tell if things were AI generated and started just...thinking more about whatever i was looking at. exploring citations rather than just looking at them, thinking about what's being said and if disparate thoughts connect to each other, trying to engage with the substance of the work. if there was little to no substance, then i could move on, and i'm sure a significant chunk of that was AI, but more importantly, it wasn't worth my time.

i genuinely and unironically hope that this is an unintended but positive outcome in a post-AI world. that people abandon attempts to determine if something is AI or not and are instead forced to think "what is there to like about this and why do i like it?"

the kids in hsm2 shouldve unionized that country club