

Yeah that’s pretty bad. We all know you can bait LLMs to spit out some evil stuff, but that they do it on their own is scary.


Yeah that’s pretty bad. We all know you can bait LLMs to spit out some evil stuff, but that they do it on their own is scary.


Ukraine is currently fighting this war for Europe too. Spending money and lifes directly in a confrontation is massively more expensive than sending weapons.


Actually this one and I just saw it on Lemmy too. As a life long Trek fan I have this quote in my head quite often.

She’s the only person on that party worth talking too…


Fuck VW. They sleep on providing cheap eCars and now the people have to suffer by losing jobs.
Availability Heuristic looks out of place. It’s pretty much the only bias I have (beside confirmation bias, which is hard to avoid as sneaky it is), but how should one survive in this world without relying on others? Without doing a scientific bias free study on every topic in life, you’re unavoidable suffering from that bias. A healthy level would be avoiding making it a rule. I regularly disagree with friends decisions, so maybe I don’t have this bias.


Dead internet, it will be drowned in bots.
One painted cow to the other: See bra!
Loneliness kills, if you’re lonely you’re probably having zero sex.
Google is doing a lot of A and B testing right now. Some days I habe to use Bing or duckduckgo, no matter what, or else I get shit results.
It’s kind of funny, because in other languages it doesn’t use this distinction and people don’t eat poisonous mushrooms because someone called them venomous by accident, or the other way around with a venomous animal.
In German venomous and poisonous is the same word. It absolutely does work.
Aside of the obvious meme joke. Well, language eveolves, maybe the distinction isn’t that important any more. Other languages don’t have it and usually you add more context to something. Also when was the last time you tried to eat an unknown animal? Or where in a situation, where you had to decide if the dangerous looking animal is only supposed to be uneatable instead of venomous?
Sieberry, it’s what Mozilla is unable to do, nice looking and working vertical tabs.
Simple Translate is nice too just mark a text and instant translation.
I still don’t care about cookies, yeah just leave me alone.
Scroll everywhere, right Click to scroll any page. Should be vanilla feature.
Keepa, for amazon price history of any product.


Your local opinion, even in a neighboring country, barely differs from another, but internationally via the internet, you are confronted with many more people. Also the internet is mostly bots by now with an agenda.
It’s really difficult to clean those data. Another case was, when they kept the markings on the training data and the result was, those who had cancer, had a doctors signature on it, so the AI could always tell the cancer from the not cancer images, going by the lack of signature. However, these people also get smarter in picking their training data, so it’s not impossible to work properly at some point.


Good. I usually go only once a month but have since refrained because Burger King, Subway and Döner are much cheaper alternatives. All McDonald’s does is being a inflation driver, driving other greedy companies to eventually also increase their prices as McDonald’s established a new normal.
Game Dev Story von Kairo Soft, it’s like 26 MB. It’s also one of the best mobile games I have ever played.
https://play.google.com/store/apps/details?id=net.kairosoft.android.gamedev3en&hl=en
Yes, there is a degeneration of replies, the longer a conversation goes. Maybe this student kind of hit the jackpot by triggering a fiction writer reply inside the dataset. It is reproducible in a similar way as the student did, by asking many questions and at a certain point you’ll notice that even simple facts get wrong. I personally have observed this with chatgpt multiple times. It’s easier to trigger by using multiple similar but non related questions, as if the AI tries to push the wider context and chat history into the same LLM training “paths” but burns them out, blocks them that way and then tries to find a different direction, similar to the path electricity from a lightning strike can take.