A first look at Midjourney 5 – a state-of-the-art generative image model that can produce photorealistic humans. We breakdown new features in v5 and discus ethical concerns surrounding AI image tech.
#ai #tech #thecodereport
💬 Chat with Me on Discord
🔗 Resources
Midjourney
AI copyright article
GPT-4 First Look
Tensorflow in 100 Seconds
🔥 Get More Content – Upgrade to PRO
Upgrade at
Use code YT25 for 25% off PRO access
🎨 My Editor Settings
– Atom One Dark
– vscode-icons
– Fira Code Font
🔖 Topics Covered
– New features in Midjourney 5
– What is Midjourney?
– Is Midjourney open sources?
– Top AI image generators
– Can AI images be hold a copyright?
– Moral and ethical concerns with AI image generators
– Midjourney 5 basics tutorial
It is March 16 2023 and you’re watching The code report but well hello my fellow Printers yesterday mid-journey released Its version 5 model in Alpha and the AI Images it can produce are shockingly Realistic like the guy with the shocked Face that got you to click on this video Is 100 artificial as a programmer and Content creator who recently lost his Job to AI I’ve been looking for a new Line of work my initial thought was to Get into modeling because my mom says I’m extremely good looking unfortunately I found out that models are now obsolete It’s now possible to generate them in All shapes and sizes and you’ll never Know the difference unless you look Really closely at the fingers there are Many different companies and projects Competing to be the best generative Image model in 2023 stable diffusion is The leading open source project then you Have tons of closed Source projects like Dolly from open Ai and countless others Trying to monetize the space but in my Opinion mid-journey is the most Impressive place to prove the images Just pop they’re vibrant realistic and Aesthetically pleasing for that we Shouldn’t be thanking mid-journey but All the photographers and artists Who Unwillingly provided the data set to Create this black magic that will make The their sons and daughters obsolete
Interestingly the U.S copyright office Just recently ruled that generative AI Art cannot be copyrighted because you Need to show proof of human authorship If you take some AI art and modify it as A human then it could become eligible But it’s reviewed on a case-by-case Basis that’s good news because it means Grifters can’t just prove something out And license it to you as a copyrighted Work the ones who will make the most Bank in this movement are the ones who Provide the models like mid-journey and Open AI open AI started as a non-profit But then transitioned to for-profit when They realized how much mother F and Money they were going to make on this Thing in any case I’m happy to pay 10 Bucks a month for co-pilot 20 bucks a Month for chat GPT and 10 bucks a month For mid-journing together they put Almost all of human creativity at your Fingertips making you a demigod in the Digital world at the same time it will Likely ruin creativity because who wants To create art when you know some Company’s just going to steal it and Remix it into an infinite number of Variations to the point where nobody can Tell the difference it takes the Incentive away from True human Talent Feast your eyes on this toad had you Told me Alex Gray painted it I’d believe You but he didn’t I’m guessing his
Artwork is in the mid-journey data set But this is my artwork I proved it at Fair and square and here’s how you can Do it too mid journey is handled Entirely in Discord there’s currently no API but I imagine that’s coming someday Once in the Discord all you have to do Is go into one of the Newbie channels or Your direct messages and type in the Imagine slash command from there just Describe whatever image you have in your Imagination and let it go to work Congratulations you can now put prompt Engineer on your resume it’ll create Four different variations of your idea Then you can choose one of them to make Even more variations or you can choose To up sample them individually currently Version 5 is an alpha and if you want to Use it use the V flag at the end of your Prompt to get highly realistic images of Humans I would also recommend adding the Q flag as 2 to increase the quality There’s all kinds of different Parameters you can use to change the Output some of the most useful ones Include aspect ratio do change the shape Of the image and chaos to control the Amount of craziness the higher the chaos The more random stuff you might get in The output image what’s especially Awesome though is that you can also Provide a starter image which can just Be a hyperlink to any image URL on the
Internet that means if you have a long Lost relative you can bring them back to Life in new artwork and photos it’s not Going to be deep fake accurate but it’s Still pretty creepy all of this Generative AI stuff might seem Terrifying if you’re a digital Creator However there is one glimmer of hope It’s very possible that you were made With AI and if you can’t tell the Difference then what does it matter this Has been the code report thanks for Watching and I will see you in the next One [Music]