br00t4c, to ai
@br00t4c@mastodon.social avatar

ScarJo v. AI, the Digital Guillotine, and Why We Play Make Believe with Misinformation

https://crooked.com/podcast/scarjo-v-ai-the-digital-guillotine-and-why-we-play-make-believe-with-misinformation/

jmaris, to ai
@jmaris@eupolicy.social avatar

The Trilemma: or why AI won't be as revolutionary as we expect.

We are in the midst of an AI revolution... or are we? As business leaders around the world scramble to integrate AI into their products and business practices, the cracks are beginning to emerge. We've seen GPT4 fail at basic reasoning, Google's AI search telling people to jump from a bridge, Microsoft's copilot inventing quotes from Vladimir Putin, and recidivism algorithms that continue to send the wrong people to prison.

markhughes, to LLMs
@markhughes@mastodon.social avatar

Using for technical design is like asking Picasso to design aircraft.

Engineer: more wings please. No, not that many, and make them symmetrical this time. No, no, no...

SenseException, to ai German
@SenseException@phpc.social avatar

Why does this remind me of an old way to rise in the ranks of search engines?

It was possible to influence search engines by having text that's colored the same as the background color in the early days of automatic indexing webpages.

If this is true, then history is repeating.
https://social.wxcafe.net/@wxcafe/112502321190198518

tansy, to ai
@tansy@wandering.shop avatar

Is anyone collecting examples of ? Because here is a particularly pernicious one.

This appeared in my YouTube suggestions as a song from Lana Del Rey's hotly anticipated Lasso album. But ut it's AI music, backed up with dozens of bot comments saying it's great.

It's not great. The voice, lyrics, melody and production are a horrible empty imitation of LDR as is the album 'photo'. It's a glimpse into how meaningless collective culture could become.

https://youtu.be/4GZbIcjBwKU?si=Lp6RMuDo3dK-RKte

Jigsaw_You, to gpt
@Jigsaw_You@mastodon.nl avatar

Soon after OpenAI released -4o on Monday, May 13, some Chinese speakers started to notice that something seemed off about this newest version of the chatbot: the tokens it uses to parse text were full of spam and porn phrases. The problem, which is likely due to inadequate data cleaning, could lead to hallucinations, poor performance, and misuse.

https://www-technologyreview-com.cdn.ampproject.org/c/s/www.technologyreview.com/2024/05/17/1092649/gpt-4o-chinese-token-polluted/amp/

marcel, to ai
@marcel@waldvogel.family avatar

Modern text generators create randomized output with no prior planning. They resist to be quality-checked by tools and processes established in the software industry.

Given this, the results are amazing. However, companies are selling the idea that these assistants will do quality checking themselves soon™.

This is mass delusion. But hey, the perks for managers/investors are worthwhile 🤷.


https://www.theverge.com/2024/5/24/24164119/google-ai-overview-mistakes-search-race-openai

bsm, to ai German
@bsm@swiss.social avatar

KI generiert kein Wissen, sondern setzt gemäss voreingestellten Parametern und Algorithmen Wörter in eine in erster Linie grammatikalisch richtige Abfolge - in neuer Zusammensetzung.

Was darüber hinausgeht, bleibt am verfügbaren Datensatz an Informationen hängen. Sind die Informationen (aka der Datensatz) Müll, dann gibt KI grammatikalisch korrekten Müll in anderer Zusammensetzung wieder.

#AI #KI #artificialintelligence #google #fail #information #disinformation

LALegault, to ai
@LALegault@newsie.social avatar

New: remote guns are tested in the West Bank on refugees:

video/mp4

billmason, to ai
@billmason@mastodon.social avatar

AI and ChatGPT now in mouse software, because of course.

https://512pixels.net/2024/04/ai-overlay-tmp-home-folder-mac-os/

KathyReid, to ai
@KathyReid@aus.social avatar

Excellent piece in #NoemaMagazine by Professor @ShannonVallor of #UniEdinburgh on the moral and experiential poverty of #AI - and what it means for us if we reduce the meaning of "human" to "producer of economic value".

A nuanced, thought-provoking and beautiful piece that argues for us to restore humanity to discussions of AI.

It provides ways to cut through the current hype cycle of #AGI and "super-human" AI, and leaves us with the fundamental question - "what does it mean to be human?".

I read it while brunching on scrambled eggs, buttered toast and hot coffee while outside on the patio, enjoying the late autumn sunshine - and I thoroughly recommend you do the same.

https://www.noemamag.com/the-danger-of-superhuman-ai-is-not-what-you-think/

jbzfn, to ai
@jbzfn@mastodon.social avatar

🇪🇺 EU Approves Historic AI Act: Pioneering Comprehensive AI Regulation | vcsi.org

"One of the key aspects of the AI Act is the enforcement mechanism, which includes substantial penalties for non-compliance. The EU Commission is vested with the authority to impose fines of up to €35 million ($38 million) or 7% of a company’s annual global revenue, whichever is higher"

https://vcsi.org/eu-ai-act-comprehensive-regulation/

strypey, to ai
@strypey@mastodon.nzoss.nz avatar

Had a chat with various MOLE* chat demos over the last few weeks. I'll be more impressed when they can run on our personal devices with no dependence on remote servers ("the cloud"). But as far as most people are concerned, we now live in the world with talking computers that can understand human speech and follow our instructions. Like in 90s Star Trek or before that, Star Wars and Blakes 7.

  • Machine Operated Learning Emulation

#AI #MOLE

noellemitchell, to Pixelfed
@noellemitchell@mstdn.social avatar

"Cox said that Emu can make “really amazing quality images” thanks to “Instagram being the data set that was used to train it” which he described as “one of the great repositories of incredible imagery.”"

This is a great reason for people to switch over to alternatives like

https://petapixel.com/2024/05/14/meta-is-using-your-instagram-photos-to-train-its-amazing-ai-image-generator/

emarktaylor, to ai
@emarktaylor@mastodon.online avatar

Via Kris @KrisSkulls
·
6h
I take it back this Google AI is awesome

emmalbriant, to ai
@emmalbriant@mastodon.online avatar

I'm just going to leave this here.

chris_hayes, to ChatGPT
@chris_hayes@fosstodon.org avatar

Recruiters using ChatGPT beware, hehehe.

Related - if anyone is looking for a web dev with agency experience working on modern headless + serverless stacks, toot in my general direction.

lopta, to ai
@lopta@mastodon.social avatar

I haven't worried about taking over the world but it just occurred to me that some countries elect truly stupid leaders, so perhaps it'll just get elected.

ErikJonker, to ai
@ErikJonker@mastodon.social avatar
WhyNotZoidberg, to ai
@WhyNotZoidberg@topspicy.social avatar

To everybody's surprise the tech bros were too engaged with the "could we" and nobody asked "should we?"

https://futurism.com/the-byte/ceo-google-ai-hallucinations

TechDesk, (edited ) to tech
@TechDesk@flipboard.social avatar

Google is taking “swift action” to remove some of the bizarre AI-generated answers appearing at the top of its search pages. “A company once known for being at the cutting edge and shipping high-quality stuff is now known for low-quality output that’s getting meme’d,” one AI founder, who wished to remain anonymous, said. Read more from @theverge.
https://flip.it/E0coKc

vsp, to ai
@vsp@mastodon.world avatar

There will be a whole mini-industry of keyboard key sticker-making in the years to come. #AI #Keyboard

In part, because of nonsense like this: https://blogs.windows.com/windowsexperience/2024/01/04/introducing-a-new-copilot-key-to-kick-off-the-year-of-ai-powered-windows-pcs/

jstatepost, to ai
@jstatepost@mstdn.social avatar

🥥 Call for venture capital!
Eye'm starting a gourmet food company that will fill the growing demand for artisanal rocks and pizza-topping glue.
Before you ask, yes, there WILL be a significant role for AI in company operations.
Interested individuals are encouraged to respond with money offers in the comments below. 🥥
, , , , , ,

jwildeboer, to ai German
@jwildeboer@social.wildeboer.net avatar

Ich bekenne mich, liebe . Ich alleine habe mit einem Shell Script auf meinem Raspberry Pi und einer von mir programmierten die ganzen Umfragen von euch, t-Online und der BILD zu eurem Thema manipuliert. Ohne kriminelle Energie. Einfach nur weil ich es kann.

preslavrachev, to ai
@preslavrachev@mastodon.social avatar

C’mon, we’ve all been using LLMs on a daily basis for more than year now. You’d think people would slowly start to get the idea. But no.

Why is everyone still using an LLM and looking for accurate answers? They are obviously not built for that.

You use an LLM to get yourself off your writer’s block, to get a few ideas how to move on, not for it to solve the end equation for you. Duh!


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