ppatel, to accessibility
@ppatel@mstdn.social avatar

Once #WWDC23 goes live, I'll have my usual thread of #accessibility related content here. I used to do this on Twitter. It involves discussing the announcements as well as testing of features announced prior to and throughout WWDC. I will welcome questions and requests for testing. I think I'll like doing this here much more than I did on Twitter. Oh and this thread will probably go on for months. I'll post things periodically as updates come.

#WWDC2023

ppatel,
@ppatel@mstdn.social avatar

If you think Apple is not interested in what everyone is calling #AI, just look at the #WWDC sessions on machine learning. They've had these sessions for years. While Microsoft, Google, and Amazon try to sell their cloud infrastructure for #ML use, Apple's already got their platform in play and has done so for those multiple years.

#GenerativeAI #MachineLearning #GPT

cwebber, to random

Please, I don't wanna hear "oops I guess you're right and that was wrong, ChatGPT must have inserted an error when it generated this statement for me" anymore

Accountability is critical in human relationships, it should be critical in our "AI" relationships too, and definitely you can't pass the buck to "well haha we don't consider that thing accountable and I used it"

shiwali,

@cwebber Taking 'AI' to court awards it an undeserved privileged status among software, adding to the hype.

#AI IS software. It is not written by software engineers but inferred from input-output pairs curated by a community of humans engineers, trainers who label or generate signals etc.

It should be held to the same scrutiny, interpretability, correctness etc.

Don't take 'AI' to court - take the companies to court who are heavily incentivized to inflate what can be done.

chrisoffner3d, to random

No mention of “AI” by Apple.

They talked about “neural networks,” “machine learning,” mentioned “encoder-decoder” architectures and “Transformers” for language prediction but did not join the AI circus. I respect that.

https://www.theverge.com/23749697/apple-wwdc-ai-machine-learning-new-features-chatbots-chatgpt

chrisoffner3d,

Tim Cook: “What people are now talking about are these Large Language Models. I think they have great promise. I do think that it’s so important to be very deliberate and very thoughtful in the development and the deployment of these because they can be so powerful that you worry about things like bias, things like misinformation, maybe worse in some cases.”

https://youtu.be/YmOC7dK3rc0
#AI #LLM #chatGPT

dtgeek, to wordpress
@dtgeek@mastodon.social avatar
dtgeek,
@dtgeek@mastodon.social avatar

AI and the block editor... yay?

At least it's a paid/optional feature. But still, a major blogging platform's newest innovation is "something to do writing for you (with mediocre results)"... sigh.

#WordPress #blogging #AI

DiplDes_AR, to gamedev

Tafi the owner of #Daz3D announces the development of a powerful text to 3D character engine. Your can generate your character via prompt and export it directly to #unity, #unreal, #blender or other popular #gameengine or #3d software. Support for open #usd will come soon. #generativeAI #AI

Link=>
https://www.prnewswire.com/news-releases/tafi-announces-development-of-powerful-text-to-3d-character-engine-301843223.html

Demo video of the generative AI based Tafi character generator.

harold, to ai
@harold@mastodon.social avatar

Keep on writing to develop better thinking skills. Don’t let the machines think for you. They may be appropriate tools in your workflow from time to time, but not for thinking.

https://jarche.com/2023/06/i-write-therefore-i-think/

#AI #ChatGPT #LLM

GambaJo, to ai German
@GambaJo@social.tchncs.de avatar
alternativeto, to ai
@alternativeto@mas.to avatar

#Instagram is reportedly testing a new feature that would allow users to chat with #AI within the app, with as many as 30 AI personalities to choose from.
https://alternativeto.net/news/2023/6/instagram-is-testing-an-ai-chatbot-feature-for-user-conversations/

AAKL, to ai

Deepmind’s AlphaDev discovers sorting algorithms that can revolutionize computing foundations #google #AI https://venturebeat.com/ai/deepminds-alphadev-discovers-sorting-algorithms-that-can-revolutionize-computing-foundations/ @VentureBeat

AAKL, to ai

This is not a toy. There need to be clear warnings of associated risks if people are going to kill themselves as a result.

"The difficulty for companies, though, is creating a bot that’s engaging and fun to talk to but doesn’t cross the line into offensive or dangerous interactions."

#Instagram is apparently testing an #AI chatbot that lets you choose from 30 personalities https://www.theverge.com/2023/6/7/23752143/instagram-ai-chatbot-feature-advice-questions-personalities-leak-screenshot @verge @james_vincent

itnewsbot, to ai

FBI warns of increasing use of AI-generated deepfakes in sextortion schemes - Enlarge

The FBI on Monday warned of the increasing use of art... - https://arstechnica.com/?p=1945970 #artificialintelligence #sextortion #deepfakes #biz#fbi #ai

pitrh, to ChatGPT
@pitrh@mastodon.social avatar

"We've seen ChatGPT generate URLs, references and even code libraries and functions that do not actually exist." https://www.infosecurity-magazine.com/news/chatgpt-spreads-malicious-packages/

Count me unsurprised.

ErikJonker, to ai
@ErikJonker@mastodon.social avatar

Interesting parts/observations in this DG Connect presentation about LLMs, not everything but you can fast-forward some parts :-) , https://www.youtube.com/live/NOOFt5Ux7g8?feature=share

teachpaperless, to ChatGPT

Just had an email exchange that left me with the distinct impression that I was communicating with #ChatGPT. The depth of conversation and the nuance in responses felt so human-like, but yet had a uncanny valley vibe - if you get what I mean?! It's truly impressive how far #AI has come - seamlessly integrating into our everyday communication. It's an exciting and a little bit surreal peek into the future of AI-driven communication. #OpenAI 🤖📧

stephenwhq, to ai
@stephenwhq@mastodon.social avatar

Good morning.
The Three Laws of AI

AI doesn't do things. Corporations will use AI to make money for shareholders. That's how we need to frame it.

AI doesn't think or feel. It's a category error to anthromorphise.

On principle, automation should lift burdens and encourage human flourishing. Stripping joy from creation is a particularly toxic use of AI.

#AI #automation #BadServantWorseMaster #WritingCommunity #AmWriting

Sherifazuhur, to ai
@Sherifazuhur@sfba.social avatar
fosslife, to ai
@fosslife@fosstodon.org avatar

2023 Future of Jobs Report from World Economic Forum says demand for AI and machine learning specialists will grow by 40 percent https://www.fosslife.org/global-outlook-ai-jobs-and-skills #AI #jobs #careers #skills #MachineLearning #DataScience #python #WEF

AAKL, to ai
WebAxe, to accessibility
@WebAxe@a11y.info avatar
itnewsbot, to ai

Healthcare Org With Over 100 Clinics Uses OpenAI's GPT-4 To Write Medical Records - US healthcare chain Carbon Health has implemented an AI tool named Carby, powered ... - https://science.slashdot.org/story/23/06/06/2155242/healthcare-org-with-over-100-clinics-uses-openais-gpt-4-to-write-medical-records?utm_source=rss1.0mainlinkanon&utm_medium=feed #ai

Imoptimal, to opensource

It's not surprising that will behave the same way did for years now - doing anything to achieve and keep for themselves, regardless of the harm to their users.

We seriously need alternatives to bull*hit, in , search and everything else that affects us all.

https://www.theverge.com/2023/6/6/23736289/microsoft-bing-chrome-search-fake-ai-chatbot

itnewsbot, to ai

OpenAI Privately Tells Developers Using Company's Tech That It Won't Compete With Them Beyond ChatGPT - Sam Altman, the erudite cofounder and chief executive of OpenAI, has a message for... - https://slashdot.org/story/23/06/07/1551250/openai-privately-tells-developers-using-companys-tech-that-it-wont-compete-with-them-beyond-chatgpt?utm_source=rss1.0mainlinkanon&utm_medium=feed #ai

kdekooter, to ai
@kdekooter@mastodon.social avatar

"Als we vandaag de computer een premier voor Nederland zouden laten genereren op basis van wat daar online over te vinden is, dan krijgen we een kunstmatig glimlachende man die de hele dag nikszeggende algemeenheden uitspuugt die ter plekke worden verzonnen."

https://www.nporadio1.nl/fragmenten/de-nieuws-bv/bc3995ac-6a88-4588-abbb-7577f06680c7/2023-06-07-druktemaker-pieter-derks-ai

itnewsbot, to ai

OpenAI Still Not Training GPT-5, Says Sam Altman - OpenAI is still not training GPT-5, months after the Microsoft-backed startup pled... - https://slashdot.org/story/23/06/07/1359239/openai-still-not-training-gpt-5-says-sam-altman?utm_source=rss1.0mainlinkanon&utm_medium=feed #ai

AAKL, to ai
  • All
  • Subscribed
  • Moderated
  • Favorites
  • JUstTest
  • mdbf
  • everett
  • osvaldo12
  • magazineikmin
  • thenastyranch
  • rosin
  • normalnudes
  • Youngstown
  • Durango
  • slotface
  • ngwrru68w68
  • kavyap
  • DreamBathrooms
  • tester
  • InstantRegret
  • ethstaker
  • GTA5RPClips
  • tacticalgear
  • Leos
  • anitta
  • modclub
  • khanakhh
  • cubers
  • cisconetworking
  • provamag3
  • megavids
  • lostlight
  • All magazines