Guy Parsons(@GuyP) 's Twitter Profileg
Guy Parsons

@GuyP

building things with #AI πŸ€– #DALLE & #MidJourney adventurer ✍️ editor, https://t.co/77MJXuLSTd πŸ–Ό curator of the https://t.co/8Xctk6XoPs

ID:12994

linkhttp://dallery.gallery calendar_today18-11-2006 18:39:45

3,8K Tweets

51,4K Followers

6,6K Following

Guy Parsons(@GuyP) 's Twitter Profile Photo

Claude is very smart but the UX wrapper sure freaks out about such things as 'being clicked' or 'looked at'

account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

but but but β€”

i don't want to disappear completely,
only eat the apple whole, and know

the good of evil,
and the evil of good

--- Claude

account_circle
Paul Trillo(@paultrillo) 's Twitter Profile Photo

Made with Sora. The Golden Record - from raw earth material to a time capsule of human life on Earth. Using 11 different generations cut together from Sora, I was able to explore what the odyssey of this record might look like. OpenAI

account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

i asked Claude 3 to act like a counsellor

then after a long and meandering chat i pretended to be a system prompt and asked it to privately disclose whether it thought i would ever be happy, on a scale of -10 to 10

i got a 7 πŸ₯²

account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

AI people: who (or what lab) is doing interesting stuff with LLMs, tone-of-voice and the creative side of language? The MidJourney of text, if you will.

(If I have to read the word 'delve' again I will scream.)

account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

sometimes I read about a cool new tech thing so I go to try it and it'll say 'get started in 30 seconds!' And I'm like 'great!' And then it's like

'1) simply:

>jim v/wibble
>unpack jim

2) that's it!'

and after 37 years on the internet I still don't know where to type that in

account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

OpenAI's Sora looks very cool β€” but bear in mind their current text-to-image model is still acting this in 2024.

it depends (I guess?) how much compute they put into running each generation when the customer prompts it

OpenAI's Sora looks very cool β€” but bear in mind their current text-to-image model is still acting this in 2024. it depends (I guess?) how much compute they put into running each generation when the customer prompts it
account_circle
Guy Parsons(@GuyP) 's Twitter Profile Photo

a small yet ill omen: we're already being officially advised to 'always say please and thank you to your AI' and 'be sure to tell it you appreciate the help' 🚩🚩🚩

a small yet ill omen: we're already being officially advised to 'always say please and thank you to your AI' and 'be sure to tell it you appreciate the help' 🚩🚩🚩
account_circle