Midjourney 5 must be stopped at all costs

Fireship
16 Mar 202303:24

TLDRThe video discusses the release of Mid-Journey's version 5 AI model, which generates highly realistic images, raising concerns about the future of modeling and the impact on creativity. It highlights the competition among generative image models, with Mid-Journey standing out for its impressive results. The video also touches on the U.S copyright office's stance on AI-generated art and the potential for AI to disrupt traditional creative incentives. Instructions on using Mid-Journey's Discord platform to create AI art are provided, emphasizing the model's capabilities and the potential for personalized image generation.

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 deceive viewers, challenging the traditional modeling industry.
  • 🌐 Various companies and projects are competing to create the best 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 capitalize on the generative AI space.
  • 🎨 Mid-journey's images are praised for their vibrancy, realism, and aesthetic appeal, crediting the vast datasets of photographers and artists.
  • 📝 The U.S. copyright office has ruled that generative AI art cannot be copyrighted without proof of human authorship.
  • 💰 OpenAI, initially a non-profit, transitioned to for-profit due to the lucrative potential of AI technology.
  • 🛠️ Mid-journey operates through Discord, offering users the ability to create images by typing commands in the Imagine slash command format.
  • 🔍 Version 5's alpha release allows users to generate highly realistic human images using the V flag and increase quality with the Q flag.
  • 🔄 Users can also provide a starter image through a hyperlink, enabling the creation of new artworks from existing images.

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 highly realistic AI-generated images.

  • How does the AI-generated image of the man with a shocked face relate to the discussion in the video script?

    -The AI-generated image of the man with a shocked face serves as an example of the impressive realism that Midjourney's version 5 model can achieve, illustrating the potential of AI in the field of image generation.

  • What is the current status of the modeling industry in light of AI advancements?

    -The modeling industry has been significantly impacted by AI advancements, as it is now possible to generate models in various shapes and sizes through AI, making traditional models potentially obsolete.

  • Which open-source project is leading the field of generative image models in 2023?

    -In 2023, Stable Diffusion is the leading open-source project in the field of generative image models.

  • What is the U.S copyright office's stance on AI-generated art?

    -The U.S copyright office has ruled that generative AI art cannot be copyrighted unless it can show proof of human authorship. AI-generated art modified by a human may become eligible for copyright on a case-by-case basis.

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

    -OpenAI started as a non-profit organization but transitioned to a for-profit model when they realized the financial potential of their AI technologies.

  • What are the subscription costs for Midjourney's services mentioned in the script?

    -The script mentions a subscription cost of $10 a month for 'co-pilot', $20 a month for 'chat GPT', and $10 a month for 'mid-journing'.

  • What is the potential concern regarding AI and human creativity?

    -The potential concern is that AI might ruin human creativity by stealing and remixing art into countless variations, reducing the incentive for true human talent and originality in creating art.

  • How can one use Midjourney's version 5 model to generate images?

    -To use Midjourney's version 5 model, one can join the Discord server, use the 'Imagine' slash command, and describe the desired image. The model will then generate four variations of the idea, which can be further refined or up-sampled individually.

  • What parameter can be used to increase the quality of images generated by Midjourney's version 5 model?

    -The 'Q' flag can be used with a value of 2 to increase the quality of images generated by Midjourney's version 5 model.

  • How does the ability to provide a starter image impact the creative process with Midjourney?

    -The ability to provide a starter image, such as a hyperlink to an image URL, allows users to incorporate existing visual elements into their AI-generated creations, enhancing the flexibility and personalization of the creative process.

Outlines

00:00

🚀 Introduction to AI and Mid-Journey's Version 5 Model

The video begins with the announcement of the release of Mid-Journey's version 5 model on March 16, 2023. The AI's ability to create highly realistic images is highlighted, using the example of a man with a shocked face that was used to attract viewers. 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 form. The video discusses various companies and projects, including open-source projects like Stable Diffusion and closed-source projects like Dolly from OpenAI, competing to be the best generative image model. The speaker expresses admiration for Mid-Journey's model for its impressive and aesthetically pleasing images, attributing its success to the vast dataset of images contributed by photographers and artists, which may inadvertently lead to the obsolescence of their children's creative endeavors.

Mindmap

Keywords

💡mid-journey

Mid-journey refers to a company mentioned in the script that has released its version 5 model in Alpha, specializing in AI-generated images. The term is significant in the video as it represents a breakthrough in AI technology, producing highly realistic images that challenge the notion of authenticity and originality in art and modeling. The video discusses the implications of mid-journey's technology on various industries, highlighting its impact on human creativity and the potential for misuse.

💡AI-generated images

AI-generated images are visual outputs created by artificial intelligence algorithms, which can mimic the style and appearance of real-world objects, people, or scenes. These images are notable for their high level of realism and variety, which can be manipulated to fit specific prompts given by users. In the context of the video, AI-generated images represent a technological advancement that raises questions about the future of human artists and the value of original creative work.

💡copyright

Copyright refers to the legal rights granted to creators of original works, including the right to reproduce, distribute, and display their work. In the video, the U.S. copyright office's ruling that generative AI art cannot be copyrighted is discussed, emphasizing the need for human authorship as a prerequisite for copyright protection. This ruling has implications for the ownership and commercialization of AI-generated content, as well as the potential for exploitation of human artists' work.

💡open source

Open source describes a type of software or content that is made publicly available, allowing users to access, use, modify, and redistribute the material without restriction. In the context of the video, the mention of the open source project Stable Diffusion highlights the collaborative nature of some AI development efforts, contrasting with closed-source projects that aim to monetize AI technology.

💡model

In the context of the video, a 'model' refers to a specific version or iteration of an AI system designed to generate images. The term is used to discuss the evolution and capabilities of AI systems like mid-journey's version 5 model, which can produce highly realistic and aesthetically pleasing images. The concept of a 'model' is central to understanding the advancements in AI technology and its impact on various industries.

💡data set

A data set is a collection of data, often used in machine learning and AI applications to train algorithms. In the context of the video, the data set is crucial for the development of AI models like mid-journey's, as it consists of images and other data that the AI uses to learn and generate new content. The video implies that the data set may include works from photographers and artists, raising questions about the ethical use of such data.

💡Discord

Discord is a communication platform designed for communities, including text, voice, and video chat. In the video, Discord is mentioned as the platform where mid-journey operates, allowing users to interact and generate AI images through the Imagine slash command. The use of Discord as a platform for AI interaction demonstrates the integration of AI technology into existing social and communication channels.

💡prompt engineering

Prompt engineering refers to the process of crafting specific and detailed instructions or prompts for AI systems to generate desired outputs. In the context of the video, prompt engineering is a skill that users of AI image generation models like mid-journey need to master in order to produce high-quality, realistic images. The concept underscores the importance of user input in the AI generation process and the creative potential it unlocks.

💡upsample

Upsampling is a process in image editing that increases the resolution of an image, often used to enhance the quality and detail of a visual. In the context of the video, upsampling is a feature of mid-journey's AI model that allows users to refine and improve the quality of the generated images, demonstrating the level of control and customization available in AI-generated content.

💡chaos parameter

The chaos parameter is a setting within AI image generation models that controls the level of randomness or unpredictability in the output image. A higher chaos value introduces more random elements, potentially leading to more creative or unexpected results. In the video, the chaos parameter is discussed as one of the various adjustable settings that users can manipulate to achieve different visual outcomes, highlighting the complexity and flexibility of AI-generated image technology.

💡starter image

A starter image is a base image or reference that users can provide to an AI image generation model to influence the style or content of the generated output. In the video, the concept of a starter image is introduced as a feature of mid-journey's AI model, allowing users to create new artwork or even 'bring back to life' images of people, such as long-lost relatives, by using a hyperlink to any image URL on the internet.

Highlights

Mid-journey released its version 5 model in Alpha with AI-generated images that are shockingly realistic.

The AI-generated image of a man with a shocked face was used to attract viewers to the video.

The speaker, a programmer and content creator, has been displaced from their job by AI and is considering a career change.

The speaker humorously suggests modeling due to their good looks, but notes that models are becoming obsolete due to generative AI.

Generative AI is advancing to the point where it can produce models in various shapes and sizes, making it hard to distinguish from real models.

There are numerous companies and projects, like Stable Diffusion and Dolly from Open AI, competing to be the best generative image model.

Mid-journey's AI is praised for creating vibrant, realistic, and aesthetically pleasing images.

The advancement of AI in art creation is attributed to the vast datasets provided by photographers and artists, which could potentially make their children's professions obsolete.

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

Grifter exploitation of AI art is limited by the case-by-case review for copyright eligibility.

The financial success of AI companies like Open AI, which transitioned from non-profit to for-profit, is highlighted.

The speaker subscribes to various AI services, including co-pilot, Chat GPT, and Mid-journey, which collectively enhance digital creativity.

AI's impact on creativity is discussed, with concerns that it may reduce the incentive for true human talent and originality.

An example of AI-generated art is given, where the speaker's artwork is mistaken for a real painting.

Mid-journey operates through Discord and offers a variety of parameters to refine the output of generated images.

The use of the 'Imagine' slash command in Discord is described for generating images with Mid-journey.

The potential for using starter images, including hyperlinks to existing images on the internet, to generate new AI art is mentioned.

The video ends with a hopeful note that digital creators might themselves have been created with AI, blurring the lines of authenticity.