|
Just as a reminder he paid $200 a user, a laughably absurd price for the site. Generously it's worth a fourth of that. Oh and currently its loosing a dollar a year per user.
Barrel Cactaur fucked around with this message at 13:06 on Oct 28, 2022 |
#
¿
Oct 28, 2022 13:03
|
|
|
|
| # ¿ May 4, 2024 02:07 |
|
|
Freakazoid_ posted:just found out about his opinion on immigration and "weird robot" may be a charitable description Thats like a pedantic anti opinion. You literally learn nothing about the position he actually holds. E:  Due to many previous rounds of  racehorses are extremely fragile physically and genetically, with Victorian breeding standards and bad incentives still causing horses to have late Hapsburg levels of inbreeding. The live cover rule in theory makes sure that you cant use samples from the #1 racer in a year to be put in every mare.
Barrel Cactaur fucked around with this message at 01:52 on Dec 18, 2022 |
|
#
¿
Dec 18, 2022 01:42
|
|
|
Nenonen posted:Also the AI doesn't want to touch sensitive subjects... except sometimes it changes its mind. And apparently language and specific wording matters a lot, I can ask for something in English and it's a no go, but once I switch to Finnish it immediately spouts out something. It's always stupid, but sometimes interesting stupid. Its running a brain-dead filter system, so of course the other languages pass, the censor only reads English.
|
|
#
¿
Jan 29, 2023 16:15
|
|
|
dr_rat posted:Oh okay, so it knows how to emulate an acid trip far better than pretty much anything else seen in tv or movies. Actually the way these ai systems work is an awful lot like an acid trip. They have a bunch of random signals that they try and recognize, which layers slight distortions that make it look more like a particular object as it shifts around. It repeats this until it's ability to recognize objects is convinced the pseudorandom blob in front of it is Sean Connery in a leprechaun outfit. It draws this and tells the world it met Sean Connery on a spirit journey.
|
|
#
¿
Mar 29, 2023 14:36
|
|
|
The machine learning business model is currently very simple. The ML companies want to scrape publicly accessible data without regard for obligations or ownership and then want to launder it through a black boxed algorithm into 100% genuine copyrighted material. Current law is suspicious of the first as a legitimate fair use, and outright hostile to the second. These are people so tone deaf they are going about proudly showing off exactly how they plan to steal peoples work and use it to replace them and wondering why this is generating economic anxiety and legal scrutiny. At the same time they are advancing technologies with major downside risks for public knowledge and public trust. I mean just in my industry I'm looking at the next generation of malware development, phishing, and a dozen other malicious secondary uses for being able to synthesize convincing human speech and text, and for rapidly prototyping software. And worse, soon the SEO blog-post in front of the recipes could become infinitely long.
|
|
#
¿
May 26, 2023 16:05
|
|
|
Arsenic Lupin posted:Reddit has found a solution to mods taking popular subreddits private. Solution: Fire the mods and get new ones! Scratch a libertarian and a fascist bleeds. It was always in service to their profits to ensure free speech, after all every speaker and listener was another ad view. Now, it has outlived its usefulness.
|
|
#
¿
Jun 14, 2023 04:21
|
|
|
Absurd Alhazred posted:He was making fun of vaccination. He would later come to regret it, although it also feels like a takedown of homeopathy, which he would not have regretted. Variolation actually, which you could somewhat make a case against because its basically "Expose them in controlled conditions and hope they don't die" Hes mocking the "pick someone who looks less sick to get scabs from" part of the procedure. If you don't have proper isolation you risk simply creating a local patient zero, because its a live virus procedure. 100% effective protection but a 3% mortality rate. Only worth it if you have news of an epidemic or are in an area with seriously endemic smallpox.
|
|
#
¿
Jul 10, 2023 13:45
|
|
|
Your off your rocker asking 500,000 for that space. At that price you could new build a 3-4000 square foot new build. That pricing is cope from someone who took a bath on commercial real estate and desperately doesn't want their portfolio to loose face value and trigger some clause in their loan. Also sounds like they have set themselves up as the HOA/condo association to ensure maximum profits.
|
|
#
¿
Jul 17, 2023 13:46
|
|
|
Tuxedo Gin posted:Japan will always win the most behind in fintech award. You got the number to fax it too and the 2500¥ fee?
|
|
#
¿
Nov 15, 2023 14:50
|
|
|
Tech VC is ruled by some of the most gullible rubes, its comical.
|
|
#
¿
Nov 18, 2023 15:55
|
|
|
As it turns out, all you need to create newspeak is a wall between talking and money. https://www.youtube.com/watch?v=eQqAY_ORhZk Fortuitously it lets you instantly ID stuff that started in tiktock but its still annoying as hell.
|
|
#
¿
Dec 8, 2023 01:36
|
|
|
Boris Galerkin posted:How about an actual article? Google turns up nothing and I’m not gonna sit here and watch a dumb video. It's a common web vulnerability, they don't invalidate or bind dynamic urls and the expiration period is very long. How "normal" sites work: you log in. In the background you take the servers public key and initiate a public private key exchange. You use your password hash (a pre shared secret) to sign a challenge from the server. You staple this to a temporary public key, send it through a SSL bridge back to the server through the servers public key, and the server validates this by finishing the SSL session. You get a url and a locally generated session cookie. This cookie just reminds your computer of its private key, it's expiration period, and what url to send to the server to tell it what temporary public key it should look back up. You can comfortably establish the shared secret through SSL, you get your email validated by a one time use token url. This is that 15 minute code. Chat GPT is alleged to just use that token url with no secondary validation that the person poking it knows the shared secret, the password. And keep it valid to connect for 7 days. This is bad because URLs are technically guessable. This is a critical security vulnerability if validated, and a violation of their PCI agreement (failure to secure data at rest). They could be in actual legal jeopardy over this. Depending on how dumb their web indexing is it could be really bad.
|
|
#
¿
Dec 11, 2023 15:12
|
|
|
The Chinese room and China brain are essentially descriptions for expert systems and neural networks respectively. The ideal way to understand why the Chinese room is not intelligent is to cast it as what it is in implementation, a machine that takes a binary code input, runs it through a set of math transformations, and gives an output. The least complex possible form of this is a logic gate, progressing to an experts system as you make the input more complex and add more rules. As the obstinate literalist noted earlier, and sufficiently complex expert systems instructions are essentially a complete manual on the task, studying the Chinese room manual long enough would eventually teach you Chinese, though in practice the best expert systems tend to have a comical error rate at tasks as open as translation. The neural net is a logical abstract of how brains work at thinking, an interconnected web of simpler systems that essentially go step by step in the process. Instead of having a truly comprehensive ruleset you have tiny pieces doing tiny things. Manually making a neural net is possible, it's how most visual programming languages work. To truly simulate brain like behavior you need a training algorithm, essentially a way of optimizing your neural net. This is still a static process for fixed neural nets. Model based neural nets add an additional step, generating a weighted map for each neuron that modifies the math that neuron is doing differently based o the tags. This is trained by having an output validator that's only good at sorting tags comparing the input request to the output for tag correlation. Additionally most of these systems work across the input itterativly, playing guess and check with their validation algorithms. So your neural net can modify it's weighing to match the tags detected in the input and re modify the output to catch localized errors. This is pretty cool, it means for example your image denoiser can match more than one exact style of painting or camera setup, and do it not just from valid images but from complete noisemaps. it's not more intelligent, just more useful. And btw, the trick to LLM remembering enough to hold a coherent conversation is just feeding past inputs and outputs into your next request. Its fairly trivial to do. For context, to simulate a human brain we currently estimate we would need 100000000000 neuron simulation neural nets, individually trained to mirror a single human neuron, consisting each of a 7 layer neural net with quite a complex input and output. Now a lot of that is taken up running the body and inefficiently remembering things, and hopefully you can make neuron templates a little bit generic. Take that as a worst case estimate, but you can see why I still call it a robot parrot. It sure did learn those words, and it sure knows how to remix them but its simply a really fancy imitation of a small part of a brain optimizing for a small task. And it still can't really learn truly new information at runtime. So essentially, do that on the fly, learning as you go, alongside every other possible human task, preferably using physical tools as well as digital ones, and maintain a coherent model of the universe based on it that you reference across tasks and maintain it for 40-80 years. Simple right? were almost there! Hopefully we obliterate capitalism before they decide the proles are obsolete enough to starve and replace with robots that are bad at their jobs but don't complain.
|
|
#
¿
Feb 2, 2024 15:50
|
|
|
Jesus III posted:I've been in a couple hundred taxis in the last two decades, all over the US and never been scammed. Maybe you all just look like rubes? Maybe dress less touristy in your daily lives or don't make up shot for internet points. I definitely got scammed going from the Tampa airport. 60% over the website estimate, no itemized receipt, cash only. The Lyft ride back to the airport when COVID ate my job was under the minimum fair for the taxis, was a nice older lady instead of some dead silence taxi man who couldn't and would explain the final cost, and almost got plasterd by someone running a residential red light at about 90
|
|
#
¿
Feb 5, 2024 21:23
|
|
|
Most wise bot adopters just use it to filter bullshit granny calls anyway, except that most bullshit granny calls come in through telephones and not text, so you just end up anoying people with another form of queue. You end up needing a human attendant to do anything you couldn't do with a click through menu on the website. Almost everyone who puts the AI in instantly gets burned.
|
|
#
¿
Feb 18, 2024 15:34
|
|
|
The Stans would rip the false idol down and ritualistically consume the producers liver. I don't get this obsession with using this tech, outside of the baseline fascist hatred of genuine art over the aesthetic of an idealized and 'purified' culture.
Barrel Cactaur fucked around with this message at 13:34 on Mar 12, 2024 |
|
#
¿
Mar 12, 2024 13:17
|
|
|
duz posted:Yes, most websites don't want to collect government ids of all their visitors and likewise the visitors don't want to give over their id to all the websites that they visit. That's what started this discussion. Like many dumb, over criminalized laws COPA is for terrible parents who won't supervise their kids. It's peak 90s lib. The party who is harmed is typically the party that put the person in harms way by giving them an unrestricted network connection. It's a massive bit of censorship and vague strict liability rules to protect parents from feeling bad their kids were on the Internet and something happened. Considering how much kids on the Internet misbehave they probably shouldn't be here at all, and the easiest way to stay in the safe harbor is to give them the 18th birthday ban.
|
|
#
¿
Mar 29, 2024 13:25
|
|
|
Tech Nightmares: Green Screening my rear end https://www.theverge.com/2024/3/27/24113838/twitch-community-update-body-part-screens-morgpie
|
|
#
¿
Mar 29, 2024 15:10
|
|
|
Electric Wrigglies posted:Neither did the old dropped bonnet on a car. It puts the hands out of the way or it loses fingats (again). That didn't have a hydraulic cylinder pulling it closed.
|
|
#
¿
Mar 30, 2024 17:32
|
|
|
greatBigJerk posted:Not saying COPPA is flawless or anything, but it is pretty good at keeping kids media companies from collecting personal data about them. It's not bad in principal, (namely I agree with a lot of what it was trying to do) but it was trying to solve a problem about 15-20 years before it happened. So it's kinda out of touch. Now I wouldn't trust entrenched interests to put in some dangerous or poisonous language. As is its better than sopa and pipa were going to be.
|
|
#
¿
Mar 31, 2024 01:25
|
|
|
PhazonLink posted:why do apps need sooo many tracking hooks for stuff? Advertisers are convinced(or more specifically, have convinced much of the ad purchasing market) that if they deliver the perfect ad to the consumer, tuned exactly to their interests, it will convince them to purchase the product. So they want to know absolutely everything about the consumer to try and create and use this magic ad targeting formula. It doesn't work and over analyzes to the point of simply showing you another version of your last 5 purchases but it's deep in the MBA/business bro hustle. Ad departments productivity is hard to measure, so they tend to be wildly over funded and like any good bureaucracy run off spend it or loose it and can't psychologically accept loosing it
|
|
#
¿
Mar 31, 2024 19:05
|
|
|
joe football posted:Maybe it's just the types of things I look for, but I think this is just real product reviews shifting to video (YouTube) more than Amazon destroying them Videos are a lot harder to spam then text with coherence, and it's way easier to spot the formulaic garbage because it typically has no dialogue or TTS dialogue.
|
|
#
¿
Apr 3, 2024 17:09
|
|
|
OddObserver posted:Have you ever seen what poor quality old-school CCTV tape video looks like? The stuff I recall seeing on TV alerts makes modem era JPEGs look pristine. It might be the killer, but it's a dead ringer for Sasquatch. Comb the woodlands!
|
|
#
¿
Apr 9, 2024 13:08
|
|
|
A the classic stable genius moves of kramering into your critics DMs and criminal tax evasion because "paying taxes would be a hardship" Also tech people who don't understand AI 200k salaries 70k servers 200k chatgpt tokens 200k free tshirt promotion Someone help me my company is dying. Barrel Cactaur fucked around with this message at 14:17 on Apr 12, 2024 |
|
#
¿
Apr 12, 2024 14:11
|
|
|
Volmarias posted:This guy is searching for the exact string "brussel sprouts air fryer temperature time" which isn't helpful. He should be writing "brussel sprouts" "air fryer" temperature time Google will helpfully silently strip that and reword your search to be about green giant brussel sprouts, your current search sponsor
|
|
#
¿
Apr 15, 2024 00:23
|
|
|
|
| # ¿ May 4, 2024 02:07 |
|
|
Volmarias posted:I still do not understand how community notes has not gotten axed. The world's most stable genius turned it off on his own posts after a day long meltdown where he got called out for spreading misinformation.
|
|
#
¿
Apr 22, 2024 19:36
|
|