True, but the principle behind the post is the beauty here. When not using the API, it costs these companies an unsustainable amount of money to make their models listen to fart sounds. I don’t use any AI myself, but I support anyone who wants to abuse the flat monthly subscription to make a company burn through money so that a plagiarism model can praise fart sounds.

https://m.youtube.com/shorts/JqvDLHshTtI

Someone recorded a voice chat to show that it says this stuff when listening.

I wrote the snippet. I mentioned why I phrased it as such elsewhere in thread:

I’ve found that people misinterpret percentage decreases as odds ratios unless I include the negative sign. The negative sign following decrease language is at best redundant and at worst actually means what is technically opposite to what I want to convey, but including it appears to make things clearer for folks for whatever reason.

Lemmy seems to be better than Reddit in this regard, given that I’ve now twice commented the explanation. Perhaps I’ll try without the negative sign in the future.

I’ve found that people misinterpret percentage decreases as odds ratios unless I include the negative sign. The negative sign following decrease language is at best redundant and at worst actually means what is technically opposite to what I want to convey, but including it appears to make things clearer for folks for whatever reason.

Pangram does work, actually. Here’s independent validation by unaffiliated scientists:

https://www.nber.org/papers/w34223

Although white papers are biased, here’s pangram’s white paper:

https://arxiv.org/pdf/2402.14873

https://www.tandfonline.com/doi/full/10.1080/10508619.2020.1729570

There are some university presses that have started their own journals in response to these issues. It’s a broken system, but no one is stopping anyone from creating new journals. All it takes is to eat away at the margins of for-profit journals by publishing fewer papers in them for them to eventually lose interest in the sector.

Only 10 states still do: https://www.kiplinger.com/taxes/states-that-still-tax-groceries

Mendelian randomization does not a causal claim justify. It’s almost certainly the case that poverty causes mental illness, but this study’s design does not permit causal inference.

Sharing what I eat would give away my identity to anyone who knows me in real life and happens to read my comment, sorry.

I’ve literally made the same exact meal for breakfast and dinner every single day for the last four years straight. Lunch changed because of an external factor, but I’ve had the same meal for lunch every day for about six months now. The meme isn’t referring to making your favorite meal, it’s referring to making the only thing you eat for that meal, ever.

Kagi has its own search engine and internal web indexing:

https://help.kagi.com/kagi/search-details/search-sources.html#search-sources

The data are promising:

https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/lithium-and-risk-for-alzheimers-disease-in-elderly-patients-with-bipolar-disorder/2F966D8BD1FF7E946C95949950AA61C1

https://www.medrxiv.org/content/10.64898/2026.01.12.26343472.abstract

As someone who worked at more than one grocery store where the manager scheduled more people per week if someone needed to constantly be on cart duty (e.g., during the winter, because folks were less likely to put their carts back during the cold), I often don’t put my cart back in the correct spot. I do so because at the stores I worked at, that would help people who want more hours be able to make a case for those hours to the manager. I often had to do so when I wanted more hours, and I was the person who did the carts. I never do this when cart duty is otherwise hard (e.g., late at night, in the cold, in the summer heat, etc.)—in those cases, I always bring my cart back inside of the store and put it completely away.

So, yes, but there are sometimes reasons to do something besides what’s courteous.

Try eating the same number of calories on a vegan diet. It’s hard. More fiber fills you quicker. The diets can’t equate both satiety and calories. Lower calories on a vegan diet is a feature, not a bug.

Recently, a company called Pangram appears to have finally made a breakthrough in this. Some studies by unaffiliated faculty (e.g., at U Chicago) have replicated its claimed false positive and false negative rates. Anecdotally, it’s the only AI detector I’ve ever run my papers through that hasn’t said my papers are written by AI.

I did some searching and found this, which seems to be a pretty good source?

https://muse.jhu.edu/article/947882

I don’t think any country should remove another country’s president under any circumstances, though.

Things appear to have changed; thanks for drawing my attention to that. I may start editing some articles in my broader area.

I can’t without doxxing myself more than I’d like. It wasn’t an article about himself, nor his research. This was about 10 years ago, so the rules may have changed. I’ll take a look and edit my post accordingly if so.

A problem with Wikipedia is that experts are not allowed to contribute to their areas of expertise because they’re “biased” (see edit below). I know a professor at a top university who used to spend his free time editing Wikipedia outside of his specific area but in his broad area of expertise as a method of disseminating science knowledge to the public. When the higher-up Wikipedia editors found out who he was, they banned his account and IP from editing.

Having the lay public write articles works when expertise isn’t required to understand something, but much of Wikipedia around science is slightly inaccurate at best. (This is still true, probably owing to the neutral point of view rule [giving weight to fringe ideas as a result] or the secondary source prioritization over primary sources.)

Edit: current Wikipedia editing rules and guidelines would not support this ban, so things appear to have changed. Wikipedia still recommends against primary sources as authoritative sources of information (recommending secondary sources instead), which is not great. But, they explicitly now welcome subject matter experts as editors.