IzzoD, to random

Well this is...about what I expected

chanezon, to random

https://bea.stollnitz.com/blog/how-gpt-works/

A clear and concise explanation of how OpenAI GPT models work by Beatriz Stollnitz.

bornach, to random
@bornach@masto.ai avatar

[Kyle Hill] on the issue of narrow AI turning out to be far narrower than we expected
https://youtu.be/l7tWoPk25yU

ramikrispin, to random
@ramikrispin@mstdn.social avatar

Great article by Chip Huyen about building LLM applications for productions 📝 👇

https://huyenchip.com/2023/04/11/llm-engineering.html

davidbisset, to opensource
@davidbisset@phpc.social avatar

ChartGPT is AI or for charts especially if you need to make charts quickly.

Free and .

https://www.chartgpt.dev/

CharlieMcHenry, to internet
@CharlieMcHenry@connectop.us avatar

I got a good chuckle from this article about - “The A.I. business influencer guys must be stopped” https://maxread.substack.com/p/the-ai-business-influencer-guys-must

nikahverse, to random

Check out this list of ways to use for :)

Source: https://www.reddit.com/r/ADHD/comments/12wuaug/ways_to_use_chat_gpt/

itnewsbot, to random
@itnewsbot@schleuss.online avatar

China Says Chatbots Must Toe the Party Line - The Communist Party outlined draft rules that would set guardrails on the rapidly growing... - https://www.nytimes.com/2023/04/24/world/asia/china-chatbots-ai.html

derthomas, to random
@derthomas@metalhead.club avatar

I caught myself using more and more at work.

This was my most recent prompt. One thing that really bothers me when using translators is that in 90% of the cases also names, label etc. are translated.

Hence my question to ChatGPT.

The result was spot on!

lamacchina, to random German
@lamacchina@swiss.social avatar

Kann ChatGPT auch Trainingspläne fürs Laufen 🏃‍♂️ erstellen?


https://www.polar.com/blog/de/laufplaene-erstellen-was-kann-chatgpt/

eldamir, to random
@eldamir@hachyderm.io avatar

Calling out on its bullshit.

"This is totally real!"
"Is it though?"
"Nah, jk lol"

scottjenson, to random
@scottjenson@social.coop avatar

Anyone have experience with https://audiopen.ai/ ?

As someone that likes to go on 'thinking walks', I frequently take short audio notes with Google Keep. It transcribes them into text but what I get is a bit jumbled and messy.

AudioPen let's me ramble into the mic for several minutes, pausing and rephrasing yet it still manages to produce a clean summary. This feels like a proper use case for ! I've been very impressed so far. Looking for any others comments on the tech.

jbzfn, to random
@jbzfn@mastodon.social avatar

「 It doesn’t bother me much that bleeding-edge ML technology sometimes gets things wrong. It bothers me a lot when it gives no warnings, cites no sources, and provides no confidence interval.

I’m unconvinced that this smoothed-out voice from nowhere deserves our attention 」
@timbray


https://www.tbray.org/ongoing/When/202x/2023/03/14/Binging

humantransit, to random
@humantransit@mastodon.online avatar

What exactly is for when it's wrong about such easily verified things? It would be so much smarter if, after doing its mysterious "language model" thing, it would just check its answer on Google. (Panama is an hour later than Costa Rica.)

CharlieMcHenry, to random
@CharlieMcHenry@connectop.us avatar

There’s always a hack… not sure which ‘internet rule’ that is, but it is so true and it appears despite the best efforts of original coders AI can be easily hacked as well. Allow me to present the ‘grandma hack’. https://www.gearrice.com/update/ai-gets-caught-up-in-feelings-and-reveals-all-its-secrets-with-this-grandma-hack/

dell, to random
@dell@journa.host avatar

Metatext doesn’t work quite as well as I remember. Any recommendations for iPhone?

lrreynolds,
@lrreynolds@mstdn.social avatar

@dell Ivory is really well done especially for someone who used for years, but Icecubes is definitely worth a try as a free app that has some really interesting features like a integration.

davidfield, to fediverse

How would you integrate into if you could?

Jigsaw_You, to programming Dutch
@Jigsaw_You@mastodon.nl avatar

, 's for , not only produces mostly insecure code but also fails to alert users to its inadequacies.

https://www.theregister.com/2023/04/21/chatgpt_insecure_code/

schizanon, to programming
@schizanon@mas.to avatar

Now that can write I think the smart career move is to get into ; defining success in a problem space was always the hard part of and will need even better tools now!

code was never the point of .

rml, to random
@rml@functional.cafe avatar

I'll become excited about when it can bootstrap on any cheap random laptop full of parts from unrelated vendors

jwildeboer, to random
@jwildeboer@social.wildeboer.net avatar

Current "AI" like is a mechanised version of the Infinite Monkey Theorem, that states that an infinite amount of monkeys and typewriters might ultimately produce the full work of Shakespeare. But definitely not any form of intelligence, IMHO. https://en.wikipedia.org/wiki/Infinite_monkey_theorem

1br0wn, to random
@1br0wn@eupolicy.social avatar

Uh-oh. Now the German DPAs have got their hands on 😏

RT @mario_gug
'The Hessen DPA also fears that OpenAI with ChatGPT also processes personal data from children, violates the data protection principles of purpose limitation and data minimization, processes the data without a legal basis (..).'
*Automated translation
https://datenschutz.hessen.de/presse/hbdi-prueft-datenverarbeitung-der-anwendung-chatgpt
https://twitter.com/mario_gug/status/1649864437560401920

itnewsbot, to random
@itnewsbot@schleuss.online avatar

AI-Powered Speaker is a Chatbot you can Actually Chat with - AI-powered chatbots are pretty cool, but most still require you to type your quest... - https://hackaday.com/2023/04/22/ai-powered-speaker-is-a-chatbot-you-can-actually-chat-with/

robotfactory, to random

Inspired by @ariadne I asked to rewrite my Mastodon bio in the style of an angry Donald Trump tweet:

"Can you believe this FAKE NEWS "cybersecurity nerd" leftist? Total disaster! Can't even make it as a YouTuber, SAD! And what's with this "non-binary" and "2nd string Disney Princess" nonsense? NO THANK YOU. Don't be fooled, folks. This loser runs some kind of roleplaying thing on Mastodon - PATHETIC. And don't get me started on this 3191220 business. Who cares?! FAKE NEWS all around. "

PrivacyDigest, to random
@PrivacyDigest@mas.to avatar
  • All
  • Subscribed
  • Moderated
  • Favorites
  • Leos
  • tsrsr
  • DreamBathrooms
  • thenastyranch
  • magazineikmin
  • hgfsjryuu7
  • Youngstown
  • InstantRegret
  • slotface
  • khanakhh
  • rosin
  • ngwrru68w68
  • kavyap
  • PowerRangers
  • normalnudes
  • tacticalgear
  • cubers
  • everett
  • vwfavf
  • ethstaker
  • osvaldo12
  • Durango
  • mdbf
  • cisconetworking
  • modclub
  • GTA5RPClips
  • tester
  • anitta
  • All magazines