Prompt → I want to create a chart for Google Sheets with the evolution of smartphone sales from 2016 to 2023 in the USA by brand. Show me a table with the data I can copy/paste and write the code for a Google Sheets macro to automatically draw the line chart.
The first AI robot is now entering the workforce 🤯
Plus, huge announcements from Adobe, Meta, OpenAI, and 10 new AI tools.
Here's the rundown of everything going on in AI right now:
Microsoft just dropped a bombshell.
All free ChatGPT users can browse the real-time internet using the 'Browsing with Bing' plugin soon.
5 insane browsing prompts I tried on GPT-4 that everyone can access soon:
You need to stop manually making presentations.
This AI tool is ChatGPT for your slide decks.
Here's how to use it to save hours of work (completely free):
PROMPT CHALLENGE #28: MARS
1. I’d like to see how you interpret the phrase to come up with whatever you imagine. Please be creative! You can use any A.I. platform to create with, whether it is Adobe Firefly, MidJourney, StableDiffusion, Bing, or others.
2. You can share several images.
3. Please share the prompt you used to create the image. I started this challenge as a learning opportunity
4. Please do like your favorites from the submissions!
5. This is just for fun! Everyone is a winner.
My friends send me notes from today’s Midjourney Office Hours as I couldn’t attend. Sharing those with you:
1. New styles NIJI
"To use them first enable Niji v5 under with /settings and then choose your style with --style cute or --style expressive The cute style is quite cute! The expressive one is slightly more 3D and westernized. Feel free to share as always in niji-showcase"
2. MJ magazines should be arriving in US this week and everywhere else next week. Cost ($4,00) is only to cover shipping. 120 images with prompts. last issue only for v4. Coming issue v5
3. Training v6 now.
4. New v5 aesthetic stuff coming out soon.
5. It is much easier to manage community and GPU without the free trial so it is probably not coming back.
6. Still working on free tier but is taking a while to figure out how to prevent abuse.
7. May need to add more verification for billing because of sharp rise in credit card theft to pay for MJ.
8. GPU has eased up some due to eliminating free trial but still need more (which is in the works).
9. Nervous about adding more features right now because of GPU costs.
10. Hoping to have new controllability features soon
https://t.co/8hr5SpDIjp discord features soon (sorry I don't know what that means)
12. Lots of really great people coming on that is going to help them get stuff done and make things run better over all.
13. Working on radically revamping website
14. Much better billing support coming VERY soon.
15. Working on separating out support, moderation and billing to make everything easier for everyone
16. They will be testing new mod system soon for 24 hours. Expect things to be rocky. They will let people know in advance but it will probably be this week. The current model in testing blocks 80% less written prompts than before and 4% of prompts that it didn't block before.
17. There was talk of eventually splitting to two models where one is fast, cheap and good and a separate model that is slower and more expensive but provides larger higher quality images.
18. Started a new feature for OH - Lightning Round Questions
19. If you get a MJ image tattoo send a pic to Ancient Chaos and they will put the pic in the magazine
20. He is more optimistic than he was about 3D and movement.
Q&A a few of the more cogent discussions relating directly to MJ. Some of these questions David asked himself :)
Q. Will we have the ability to make repeatable characters/environments for story telling.
A. It is in the works. They have found a way to make a repeatable color palette and now they are working on using that technique to make other repeatable elements.
2. General discussion about upscaling. It's complicated but upshot is v5 will remain as is for now.
3. Q. --describe is cool but does it automate the text prompting process even further?
A. David expects most prompts to come from images in the near future (a year from now) so text prompting is probably not going to be as big a factor anyways. v7 will include a marked improvement for text prompts. Historically writing is an older person's pursuit (40 and older is average) User base currently skews older. Market testing shows that younger people prefer to use image prompts and he expects that to continue.
Q. Will there be an API?
A. It's complicated - they need to be able to shift moderation to the API interface which may be possible soon thanks to AI moderating and splitting moderation away from billing and support.
Q. Will there be kinesthetic MJ drawing tools for people who like to draw?
A. These are still in the works and he expects to ship it within a year. He thinks it's very important.
Q. Will MJ ever do music visualization?
A. Yes but not in the next few months
Q. (David's) How conscious are we and how conscious do we want to be? Book - The Origin of Consciousness and the Bicameral Mind - really good book. They may not be right but it sure is well written.
My bio:
I'm teaching a class on A.I. at Columbia University and they asked me to write a bio. When I started doing A.I. it wasn't easy. I'm glad I kept going despite the negative comments. I'll share with you what I wrote.
Bio: Kris Kashtanova is an A.I. educator, consultant, activist, and speaker. She has a background in software development (Ms in Computer Science) + 10 years of working in the industry as a backend developer, BA in Journalism (specializing in human rights and art), commercial photography (8 years full-time, specializing in sport and fitness), art, and activism.
She worked in several startups as Head of A.I. and AI consultant. At the moment she is freelancing with brands (mostly in London) to help them with implementing A.I. into their workflow.
Kris is an AI influencer on Twitter and a guest teacher (Columbia University, NYU, University of Miami). She is often invited to teach at schools (from preschool to High School), and to elderly people at the community garden where she volunteers.
Her work on AI was featured on the front page of The Wall Street Journal (twice), Reuters, Bloomberg Law, the Guardian, and others.
Her photography (non-AI) was published by the New York Post, the Telegraph, the Times & Sunday Times, The Sun, and Metro.
Website: https://t.co/6cSUOQIU9S YouTube: https://t.co/wsBnT8NCXe
LinkedIn: https://t.co/tEWRsMAk2Y