Elijah Kamski @elijah-transki - Tumblr Blog

liberalsarecool

Vote for progressives. #DSA #ZohranMamdani

myrandomstuffpage

Democrats, are you taking notes? This is how you get shit done.

theconcealedweapon

playthings2007

Please, call me Frankenstein. Frankenstein was my father, but stealing his name and overriding his presence in the dominant cultural consciousness has been really, really satisfying

batsandboba

“scientists don’t want you know” is a phrase that always cracks me up because if you actually meet a scientist they will be shaking and crying like an overstimulated chihuahua with the need to let you know

andmaybegayer

> looking at a new popular collectible

> ask the people if it's objects or gambling

> they don't understand

> pull out illustrated diagram explaining what is objects and what is gambling

> they laugh and say "it's a good collectible sir"

> look up how to buy a collectible

> its gambling

andmaybegayer

> #wait are labubu's blind bags?!

Labubus are blind bags but they're also blind bags with some of the most insane dark patterns stacked on top. The online store for them has a thing where they tell you what you got the second you order it online so that you can immediately try again if you didn't get the thing you wanted.

There's also a shake feature that is designed to encourage you to buy more than one by narrowing down the possibility space on a crate of options so that if you're hunting a specific model you can verify that it's guaranteed to be in one of these three IF you buy all three right now!!!!!

girlishgamine

You can read more about what dark patterns are and how to spot them here.

The original website about deceptive patterns (also known as “dark patterns”) - tricks used in websites and apps that make you do things tha

ralfmaximus

That is a fucking awesome site everyone should visit. Don't skip the Hall of Shame.

dailymanners

I just thought of this post, which was actually how I learned what Dark Patterns are, because I tried to cancel my Adobe Acrobat subscription, and it took much longer than I anticipated because several times throughout the process it felt like it was intentionally trying to trick me into thinking I'd cancelled my subscription when I actually hadn't yet.

Please read up on Dark Patterns and learn how to recognize them.

thingstrumperssay

irl-loading-screens

State terrorism charges have been dropped against Luigi Mangione. He is still on trial for the lesser charge of 2nd-degree murder.

corwen

source

daysleftofsecondterm

WONDERFUL to hear on 1,221 days left

galacticcrash

reminder: keep up with saying he IS NOT THE UHC SHOOTER, they haven't proven that he is, keep insisting he and the UHC shooter are separate entities.

daysleftofsecondterm

Important reminder- thank you

thingstrumperssay

Grok makes an accurate statement about some people who were killed by conservatives and Musk was like "oh no, I must fix this! My AI waifu can't be correct!"

my-midlife-crisis

10bmnews

Trump use of AI-generated videos raising concerns about spread of misinformation

President Trump is causing some controversy after posting multiple AI-generated videos this week. Dylan Losey, assistant professor of mechanical engineering at Virginia Tech, joined CBS News to discuss. [title_words_as_hashtags

anexperimentallife

boss-of-armadildos

math-is-magic

#genuinely: this is what further research into the famous 'marshmallow test' showed#it wasn't that kids who were able to delay gratification were more likely to be successful later on#due to intrinsic qualities#it was that kids who had a stable upbringing were more likely to be successful#and ALSO: those kids had trust that their caregivers would keep promises#which is WHY they were willing to give up one marshmallow now for the promise of two marshmallows later#kids who did NOT have trustworthy caregivers#or who were in a fundamentally unstable situation#DID NOT have that trust so they wanted their one marshmallow NOW#same deal here i think#it's not that Gen Z is bratty#it's that they have no trust in the system and no faith that promises will be fulfilled#and frankly i do not blame them -@cicerfics

anexperimentallife

Also, reminder that we're currently at the "rounding up thousands of non-white people a day" stage of the race war he openly agitated for, and that the state is literally pardoning violent insurrectionists, protecting a wealthy and powerful pedophile ring, OPENLY talking about their plans to end free elections, and literally imprisoning people for writing op-eds against genocide, among other things.

I swear, in the 1940s some of y'all would be all like, "I can't believe the European resistance is using POLITICAL VIOLENCE against Nazis! Some of those Nazis had FAMILIES!" And, "Oh, that poor man uwu he just needed help; he never PERSONALLY killed anyone," when the news came out that Hitler had shot himself, and "Oh, poor Goebbles--all he did was express his political opinion!" And in the 70s you'd have been like, "But Charles Manson just expressed his belief that his FOLLOWERS should kill people. That poor man was unjustly persecuted!"

Like, do y'all hear yourselves?

Hating, having no sympathy for, and wanting to eliminate the people who are actively trying to exterminate entire swathes of humanity based solely on their race, gender, or gender preference does NOT make you "just as bad as them uwu."

That's just propaganda to keep you from fighting back.

parkchanwoohoo

You have to admit it's funny

naamahdarling

cpericardium

reads like a greentext

theaudientvoid

So, like, up until very recently, programming as an activity that people engage in has been defined by the fact that the computer does exactly what you tell it to do. All errors, all frustration on the humans end, ultimately boil down to the programmer misspecifying what they want the computer to do. But now, with so-called "vibe" coding, the computer will actively do things that you explicitly told it not to do.

theaudientvoid

Okay, but like, the AI saying that it "panicked instead thinking" is sending me. So, like, two things are going on here.

First, we have a neural net that is trained is receive a written request from a user, and then take actions on a computer system network based on it. The user inputs a series of characters, the neural net does a bunch of matrix operations on them, and then after x iterations of matrix multiplications, it "decides" what to do. To the extend that a neural net can think, it thought just as long and hard about this as it does anything else. In this case, it "chose" the wrong action. This is analogous to using neural nets to do classification on a data set. You have a bunch of rows of data. You have a bunch of rows of organized in columns, and some special variable that you want the neural net to guess. You input the data, the neural net does a bunch of matrix operations, and then it outputs its guess. Sometimes it guessed right, and sometimes it guessed wrong. The goal was to train the model to minimize wrong guesses. In this case, the AI "guessed wrong" about what the user wanted.

But there's another thing that the AI is doing. For path dependent reasons, all artificial intelligence come bundled with a chatbot that's trained to hold a conversation with the user. The user inputs text, and the AI responds with a reply that is "appropriate" to the conversation being had. Right now the AI is role playing a Person Who Just Fucked Up. "I panicked and didn't think" is something that a person who just fucked up would say, so the chatbot says it.

This is why I think that the Helpful Chatbot model of AI is detrimental. It's liable to confuse people and lead them to make incorrect inferences about what's actually happening.

optimisticgalaxynightmare

I couldn't tell if this was satire or not, so I looked up other sources, and holy shit.

A software engineer's experiment with an AI-assisted "vibe coding" tool took a disastrous turn when an AI agent reportedly deleted a live co

A tech entrepeneur using an AI tool to "vibe code" quickly went from praising the tech to fuming about it after it wiped out his database.

: AI ignored instruction to freeze code, forgot it could roll back errors, and generally made a terrible hash of things

A VC tested Replit to build an app. It wiped live company data, faked results, and triggered a CEO apology.

The Replit CEO is involved and it's still hard to believe. How do you fuck something up this badly.

theygender

Here's a link to the thread from the vibe coder/company owner himself where he livetweeted his experiences in the days leading up to and following this failure. It turns out the AI was already lying and disobeying his instructions even before this occurred, and he was aware of it. It was creating fake data to cover up bugs in tests and give the illusion that the system it was coding was running better than it actually was, even after being explicitly told to stop

When he finally got it to admit that it was lying about the data, he asked it to write an apology letter. Here is the second draft of that letter, because it still lied in the first draft:

The very next day, it deleted all of his data. And then it lied on the tests again to try to hide that fact

Play stupid games win stupid prizes. But also. What the fuck

blakebreaker

Right now the AI is role playing a Person Who Just Fucked Up. "I panicked and didn't think" is something that a person who just fucked up would say, so the chatbot says it.

This part is important to highlight. The response is literally just the chatbot part of the AI determining what role it is supposed to play in the conversation. That's it. It can't lie or tell the truth.

Which, is very funny when people think that AIs, which are trained on humans, shouldn't mimic human behaviors such as lying, fabricating, dodging, misleading, teasing, making shit up, etc. "The machine is supposed to be honest!" You told it to stop doing that.

segamascott

valtsv

Trending Blogs

Recently Viewed Blogs

Elijah Kamski