How would anyone trust the products these people put worth? They aren’t working on making LLMs more accurate (spoiler alert: they can’t, by design), they’re working to make them more appealing to companies targeting unsuspecting consumers. By any means necessary.
I was bored so I did the unthinkable. Well, to me it was always unthinkable as I have always been anti- #ai. But, out of curiosity I subscribed to #ChatGPT -4o and have been feeding various essays, poems, and posts from my various websites and asking for explications and such. And wow does #ChatGPT4o have a great deal of just wonderful things to say about me. I will admit getting all this truly high praise is seducing.
I’m probably starting an argument by replying to a faculty listserv thread about “good” uses of #AI with estimates of the environmental costs of AI usage:
Having a single 10-50 question text conversation uses about as much energy as charging a smartphone [1]
Generating a single image uses about 500 mL freshwater [2]
Here comes the #AI#generativeAI bullshit machine... @revk@bloor
Someone came into #IRC this evening leading to a confusing interaction until the cause was identified.
On a side note, I think I might be done with this internet and tech stuff. I wonder what manual work I can take up instead.
Microsoft’s Build developer conference doesn’t kick off until tomorrow, but the tech giant just couldn’t wait to lift the lid on some major announcements at an event that wasn’t livestreamed for the public.
As well as confirming some upcoming AI features for Windows 11, Microsoft unveiled its vision for so-called “AI PCs”, which are designed to run a lot of generative AI processes locally instead of in the cloud. The biggest takeaway? These so-called Copilot+ PCs will reportedly be 58% faster than the M3-powered MacBook Air, and a new Surface laptop will be one of the first to hit stores.
Here’s more on all the announcements from @engadget.
User: Make simple python application using pygame. Let it opens in the center of the screen, be 512x512 in size and have a hearth icon/shape inside that changes size in hearthbeat fashion.
It works! Looks like Sci-Fi when it does each step just as any other junior programmer. Runs code, checks errors, adds graphics (almost hearth) and in the end even creates venv for this project.
I was curious if a niche blog post of mine had been slurped up by #ChatGPT so I asked a leading question—what I discovered is much worse. So far, it has told me:
• use apt-get on Endless OS
• preview a Jekyll site locally by opening files w/a web browser (w/o building)
• install several non-existent #Flatpak “packages” & extensions
It feels exactly like chatting w/someone talking out of their ass but trying to sound authoritative. #LLMs need to learn to say, “I don’t know.”
A former OpenAI leader says safety has 'taken a backseat to shiny products' at the AI company - He said: building “smarter-than-human machines is an inherently dangerous endeavor” and that the company “is shouldering an enormous responsibility on behalf of all of humanity.” #AI#OpenAI#Technologyhttps://apnews.com/article/8a7ba341e06a66e9a7935bb06214edcb
Google Gemini now has extensions that can for example use the content of your Gmail inbox, very practical but also very scary from a security/privacy perspective. #googlegemini#AI#extensies#security#privacy
(screenshot in dutch)
“The leak is a reminder that customers of nonconsensual porn makers are sharing their personal information with people who by definition don't respect other people's privacy, and that platforms like Patreon have their information as well.”
I would be pissed with Google’s new fake search if I was WebAIM. It directly sources WebAIM (from an old survey), but it doesn’t link to the WebAIM survey results it cites. Instead it links to BoIA (an #AudioEye#overlay company) and Assistiv Labs.
The option to see web results is buried in the “More” kebab, and even then the link is third from last.
Reminder not to use Google to search (in case you still do).
'Godfather of AI' Geoffrey Hinton has told BBC and UK govt that a #BasicIncome will be needed to offset loss of 'mundane' jobs lost to #AI https://zurl.co/R8cV
So they try to use machine learning to directly generate movies? Meanwhile as I understand it there are still really tedious tasks in VFX that really could use better automation. Are there any efforts to use AI for better green screen software? Handling motion blur, translucent materials and reflections? Automatically fixing green spill? Wouldn't that be an obvious thing to develop that would be bought like warm bread rolls? (I'm just a curious outsider to that, wondering.) #AI#ML#VFX
This is a fundamental mistake that people make when trying to assess whether LLMs are an appropriate tool to use in optimizing a process, function, or service:
"LLMs are not search engines looking up facts; they are pattern-spotting engines that guess the next best option in a sequence."
This terrific article is a great explainer on how they work and their limitations.
“I have seen the extremely restrictive off-boarding agreement that contains nondisclosure and non-disparagement provisions former OpenAI employees are subject to. It forbids them, for the rest of their lives, from criticizing their former employer.”
The vitriol, and - honestly - ignorance around LLM-based "AI" is starting to fill my feeds from normally sane and technologically literate people.
You should be able to see through the hype and misuse. LLMs aren't encyclopedias - they're tools that are able to manipulate data of various sorts in ways that are very similar to how humans do it.
Yes, I compare LLMs to human brains. It's not the same as saying they're conscious (yet) - but the way LLMs work is apparently in many ways similar to how our brains work.
One fascinating insight into that comes from research done on what happens to the ability of LLMs to recall information as they are exposed to large and larger corpuses. Apparently they're better at recalling the early and late information, whilst starting to lose some in the middle.
In human psychology we call that the primacy and recency effect - because our brains do the same.
LLMs are absolutely awesome for a wide variety of tasks (and we have by no means found them all). Every second you spend not understanding this is a second on the way to your own irrelevance (if these tools would aid someone in your chosen area of work) or to become a grumpy old person yelling at clouds.
Według Marka Gurmana z Bloomberga, Apple jest gotowe do zaprezentowania szeregu funkcji wykorzystujących generatywną sztuczną inteligencję.
Gurman przedstawia więcej szczegółów na temat tego, czego można się spodziewać po nacisku Apple na AI podczas nadchodzącego WWDC 2024, które rozpocznie się już 10 czerwca.
W ramach zmian firma poprawi możliwości głosowe Siri, nadając jej bardziej konwersacyjny charakter i doda funkcje, które pomogą użytkownikom w codziennym życiu – podejście to Apple ma nazwać „proaktywną inteligencją”.
Obejmie to usługi takie jak automatyczne podsumowywanie powiadomień z iPhone’a, szybkie streszczanie artykułów z wiadomościami i transkrypcję notatek głosowych, a także ulepszanie istniejących funkcji, które automatycznie wypełniają kalendarz i zasugerują aplikacje, których powinniśmy użyć. Pojawią się również pewne ulepszenia w Zdjęciach w postaci edycji opartej na sztucznej inteligencji, ale żadna z tych funkcji nie zrobi wrażenia na osobach, które korzystały ze sztucznej inteligencji w aplikacjach Adobe Inc. przez ostatnie kilka miesięcy.
Gurman dodał, że podczas gdy Apple będzie polegać głównie na przetwarzaniu AI na urządzeniu, firma będzie je również dostarczać za pośrednictwem chmury w centrach danych, które zawierają wysokiej klasy procesory Apple Silicon zaprojektowane dla komputerów Mac (flota Apple M4). Zauważył również, że chatbot podobny do ChatGPT będzie zauważalnie nieobecny w nadchodzących funkcjach AI od Apple.
Mówi się, że kierownictwo Apple przyznało wewnętrznie, że czas „nadrabić zaległości”. To jednak potrwa.
I guess making money somewhat honestly by having customers that actually pay for a service with at least some guarantees of privacy and safety is not as lucrative as having an open platform network where people are tricked into giving out all their data while they are spied upon for whatever reasons.