Midjourney 5 must be stopped at all costs

Fireship
16 Mar 202303:24

TLDRIn the video, the host discusses the release of Mid-Journey's version 5 AI model, which generates hyper-realistic images. The impact of AI on jobs, particularly in content creation and modeling, is highlighted. The video also touches on the legal aspects of AI-generated art, mentioning the U.S. copyright office's stance. The host shares their experience with Mid-Journey's platform, explaining how it operates via Discord and the various commands and parameters available to users for creating unique images. The video concludes with a contemplation on the future of digital creation in the face of advancing AI technology.

Takeaways

  • 🚀 Mid-journey has released its version 5 model in Alpha, producing highly realistic AI images.
  • 😲 The AI-generated images are so lifelike that they can replace models, indicating a shift in the industry.
  • 🎨 Various companies and projects are competing to be the best in generative image models, with Stable Diffusion leading as an open-source project.
  • 🏢 Closed-source projects like Dolly from OpenAI are among the many trying to monetize the generative AI space.
  • 🌟 The speaker praises the vibrancy and realism of Mid-journey's images, crediting the photographers and artists whose work contributed to the AI's capabilities.
  • 📝 The U.S. copyright office has ruled that generative AI art cannot be copyrighted without proof of human authorship.
  • 💰 The entities that stand to profit most are the model providers, such as Mid-journey and OpenAI, which has transitioned from non-profit to for-profit.
  • 🛠️ The speaker discusses the ease of using Mid-journey within Discord, highlighting its potential for an API in the future.
  • 🎨 Mid-journey's version 5 allows users to generate images with various parameters, including aspect ratio and 'chaos' level.
  • 🔍 Users can provide a starter image via a hyperlink, enabling the recreation of people or things in new artwork and photos.
  • 🌐 While generative AI may be intimidating for digital creators, there is hope that AI could also be the creator, blurring the lines of authorship.

Q & A

  • What significant update did Midjourney release on March 16, 2023?

    -On March 16, 2023, Midjourney released its version 5 model in Alpha, which is capable of producing shockingly realistic AI images.

  • How does the speaker's experience with losing their job to AI influence their perspective on generative image models?

    -The speaker, being a programmer and content creator who lost their job to AI, initially considered a career in modeling due to their good looks. However, they discovered that models are becoming obsolete due to the advanced capabilities of generative image models, which can create realistic images of any shape and size.

  • What is the current stance of the U.S copyright office regarding generative AI art?

    -The U.S copyright office has recently ruled that generative AI art cannot be copyrighted because proof of human authorship is required. AI art can only be considered for copyright if it is modified by a human.

  • How has OpenAI's status evolved since its inception?

    -OpenAI started as a non-profit organization but transitioned to a for-profit entity when they realized the potential for significant financial gain from their AI technologies.

  • What are the speaker's thoughts on the impact of AI on human creativity?

    -The speaker expresses concern that AI might ruin human creativity. They argue that with companies able to steal and remix AI-generated art into countless variations, the incentive for true human talent and originality may be diminished.

  • How can one currently use Midjourney's version 5 model?

    -Midjourney's version 5 model can be used by joining their Discord server and using the 'Imagine' slash command followed by a descriptive text prompt. The model generates four variations of the imagined image, which can be further refined or up-sampled for higher quality.

  • What is the significance of the 'V' flag when using Midjourney's version 5 model?

    -The 'V' flag, when used at the end of a prompt in Midjourney's version 5 model, instructs the AI to produce highly realistic images of humans.

  • What does the 'Q' flag do in Midjourney's version 5?

    -The 'Q' flag in Midjourney's version 5 is used to increase the quality of the generated images. A value of '2' is recommended for better quality outputs.

  • How can a starter image be provided to Midjourney's AI for generating new artwork?

    -A starter image can be provided as a hyperlink to any image URL on the internet. This allows users to incorporate existing images into new AI-generated artwork, such as bringing a long-lost relative back to life in a new piece of art or photo.

  • What is the speaker's opinion on the potential future for digital creators in the context of generative AI?

    -The speaker suggests that there is hope for digital creators, as it is possible that they themselves were created with AI. If the distinction between human and AI-generated art becomes indistinguishable, the speaker implies that the importance of the creator's origin may lessen.

  • What are some parameters that can be adjusted in Midjourney to change the output image?

    -In Midjourney, parameters such as aspect ratio can change the shape of the image, and chaos can control the amount of randomness in the output. Higher chaos levels introduce more unexpected elements into the generated image.

Outlines

00:00

🚀 Introduction to Mid-Journey's AI Model

The video begins with the announcement of the release of Mid-Journey's version 5 model in Alpha on March 16, 2023. The focus is on the AI's ability to create incredibly realistic images, as exemplified by the thumbnail featuring a person with a shocked face. The speaker, a programmer and content creator who lost his job to AI, humorously considers a career in modeling before acknowledging the obsolescence of human models due to AI's capability to generate any desired appearance. The video introduces various companies and projects, such as Stable Diffusion and Dolly from Open AI, competing to develop the best generative image model. The speaker expresses admiration for Mid-Journey's model for its vibrant, realistic, and aesthetically pleasing images, crediting the photographers and artists whose work has been used to train the AI. The discussion then shifts to the U.S. copyright office's recent ruling that generative AI art cannot be copyrighted unless it can be proven to have human authorship.

Mindmap

Keywords

💡mid-journey

Mid-journey is a reference to a company that has released a version 5 model in Alpha, which is an AI system capable of generating highly realistic images. In the context of the video, it represents the advancement of AI technology in the field of image generation, and its impact on various industries, including content creation and modeling. The speaker mentions mid-journey as an impressive platform that can create aesthetically pleasing images, suggesting a significant role in the current AI landscape.

💡AI images

AI images refer to visual content that is artificially generated by AI systems, like the one described in the video by mid-journey. These images are created using complex algorithms and machine learning models, which can produce results that are shockingly realistic. The video emphasizes the high quality of AI-generated images, to the point where they can mimic human-generated content, which raises questions about the future of human creativity and the value of original artwork.

💡 Generative image model

A generative image model is a type of AI system that is designed to create new images from scratch based on patterns and features it has learned from existing datasets. In the video, the generative models are highlighted as a significant technological advancement, with various companies and projects competing to develop the best models. These models are capable of generating images in all shapes and sizes, which poses a challenge to traditional professions like modeling and art creation.

💡Stable diffusion

Stable diffusion is mentioned as the leading open-source project in the generative AI image model domain. It represents a collaborative effort to advance AI technology and make it accessible to a broader audience. The video positions stable diffusion as a key player in the competition among different AI projects, emphasizing the importance of open-source initiatives in driving innovation and democratizing access to cutting-edge technology.

💡Dolly

Dolly is cited as an example of a closed-source project developed by Open AI, which is competing in the generative AI space. The mention of Dolly illustrates the diversity of approaches to AI development, with some companies opting for a closed-source model that may focus on monetization. This contrasts with open-source projects like stable diffusion, highlighting the different strategies companies use to capitalize on AI technology.

💡Copyright

Copyright is a legal concept discussed in the video in relation to AI-generated art. It explains that the U.S. copyright office has ruled generative AI art cannot be copyrighted unless human authorship can be proven. This ruling has implications for the ownership and distribution of AI-generated content, potentially allowing for wider use and modification of such art without legal restrictions. The video uses this to discuss the broader implications for creators and the value of original human creativity.

💡Co-pilot

Co-pilot is mentioned as a service provided by Open AI, for which the speaker is willing to pay a monthly fee. It represents the commercialization of AI technology and the willingness of users to subscribe to such services to enhance their productivity or creativity. The reference to co-pilot in the video underscores the integration of AI tools into various aspects of professional and creative work, and the potential financial success for companies offering these AI services.

💡Chat GPT

Chat GPT is another AI service mentioned in the video, which the speaker also subscribes to. It likely refers to an AI-powered chatbot or conversational agent that can assist users in various tasks, from generating text to providing information. The mention of Chat GPT illustrates the diverse applications of AI beyond image generation, and how AI services can become integral parts of a user's digital toolkit, enhancing their capabilities in various domains.

💡Mid-journeying

The term 'mid-journeying' is used to describe the experience of using the mid-journey AI platform. It involves interacting with the AI through Discord and utilizing the 'Imagine' command to generate images based on user prompts. The concept highlights the participatory nature of AI engagement, where users actively contribute to the creative process by providing inputs that the AI then transforms into visual outputs.

💡V flag

The 'V flag' is a specific parameter used within the mid-journey AI platform to generate highly realistic images of humans. It is an example of the various technical tools and settings available to users to refine and customize their AI-generated content. The video uses the V flag to illustrate the level of control and specificity that users can have over the AI's output, emphasizing the platform's capabilities and the potential for high-quality results.

💡Q flag

The 'Q flag' is a quality enhancement parameter used in the mid-journey AI platform. By setting it to 2, users can increase the quality of their generated images. This keyword showcases the platform's ability to adjust the resolution and detail of the AI-generated content, allowing users to achieve their desired level of visual fidelity. The video highlights the importance of such customization options in creating content that meets users' expectations and needs.

💡Starter image

A 'starter image' refers to an initial image or a hyperlink to an image URL that users can provide to the AI to influence the generation process. This concept is showcased in the video as a way to bring personal elements into the AI's creative process, such as reviving a long-lost relative in new artwork. The starter image serves as a creative springboard, blending AI's generative capabilities with user-provided content, and results in a unique fusion of AI and human input.

Highlights

Mid-journey releases its version 5 model in Alpha, producing shockingly realistic AI images.

The AI-generated images are so realistic that they can make models obsolete.

Numerous companies and projects are competing to create the best generative image model in 2023.

Stable diffusion is the leading open-source project for generative AI images.

Open AI's Dolly and other closed-source projects are monetizing generative AI.

Mid-journey's images are vibrant, realistic, and aesthetically pleasing.

Photographers and artists have unwillingly contributed to the data sets for these AI models.

U.S copyright office ruled generative AI art cannot be copyrighted without proof of human authorship.

Open AI transitioned from non-profit to for-profit due to the potential earnings from AI technology.

The subscription model allows access to powerful AI tools like co-pilot, chat GPT, and mid-journey.

AI tools put human creativity at users' fingertips, potentially making them digital demigods.

The ease of AI-generated art may diminish the incentive for true human talent and creativity.

Mid-journey operates on Discord with no current API, but one may be coming in the future.

The Imagine/slash command in Discord allows users to create AI-generated images from text prompts.

Version 5 of mid-journey, when used with the V flag, produces highly realistic human images.

The Q flag as 2 increases the quality of the generated images.

Parameters like aspect ratio and chaos can be adjusted to change the output image.

Starter images can be provided as hyperlinks to any image URL on the internet.

Generative AI can bring long-lost relatives back to life in new artwork and photos.

There is hope for digital creators as AI-generated content may not be distinguishable from human-made.

Casual Browsing

The Economy Can't Be Stopped | Real Estate, Stocks, Gold, Bitcoin

2024-04-06 02:00:01

Why I Stopped Recommending Printify & Stopped Making Kittl Tutorials

2024-04-22 09:35:00

Top 5 Must Have Stable Diffusion Extension

2024-05-17 08:15:02

They all Cheated at the Off Road Games

2024-03-29 12:25:01

Why I STOPPED Using ChatGPT...

2024-04-11 19:20:00

Why I STOPPED Using ChatGPT...

2024-04-11 19:20:01