BestAI Newsroom research note

This evergreen history article uses authoritative archives and official records. Exact dates are used when documented; gradual inventions and rollouts are described as periods rather than being assigned a misleading single birthday.

Quick facts

  • Midjourney was founded by David Holz as an independent research laboratory.
  • The service entered public beta in 2022 and initially operated mainly through commands inside Discord.
  • Its models became known for polished, cinematic and illustrative images from text prompts.
  • Rapid version updates improved anatomy, text rendering, prompt understanding, editing and personalization.
  • Midjourney later expanded its website tools and moved from still images toward video generation.

An independent research laboratory

David Holz previously co-founded Leap Motion, a company known for hand-tracking technology. He later formed Midjourney as a small, self-funded research organization exploring new ways to expand human imagination.

Rather than launching through a conventional design application, Midjourney used Discord. Users typed commands in shared channels, watched images appear and learned from one another’s prompts. This public workflow made the community part of the product’s growth.

The 2022 public beta

Midjourney opened its public beta in 2022 during the first major wave of modern text-to-image systems. Early outputs were surreal and inconsistent, but they already displayed a recognizable aesthetic. Frequent model versions quickly improved faces, composition and prompt interpretation.

The service required paid subscriptions after limited trials. Images generated in public channels could be seen and remixed by others, encouraging experimentation while creating privacy concerns for commercial work.

A new visual language

Midjourney became popular among concept artists, marketers, filmmakers, architects and hobbyists. Prompts could describe camera lenses, lighting, materials, art movements and cinematic moods. The model often produced a polished result even from relatively short instructions.

That strength created criticism that AI images could become visually repetitive or substitute surface style for deliberate design. Skilled users responded by developing prompt systems, reference images, editing workflows and external compositing.

Copyright, artists and synthetic media

Artists and rights holders questioned whether training data had been licensed and whether generated images could imitate living creators. Lawsuits and policy debates challenged multiple generative-image companies, including Midjourney. The legal outcome varies by jurisdiction and remains unsettled.

The platform’s public nature also allowed realistic misinformation, impersonation and misleading imagery. Rules, moderation and restrictions were introduced, but generative media continued to test the boundary between creative freedom and harm prevention.

Web creation, personalization and video

Midjourney developed a fuller website for prompting, browsing, organizing and editing. Features such as image prompts, style references, character references, inpainting, zoom, personalization and moodboards gave users more control than the original Discord command workflow.

The move into video expanded Midjourney’s ambition from creating a single image to generating motion and scenes. Its historical significance lies in showing that a small independent laboratory could build one of the most culturally influential generative-media products.

Common misconceptions

  • Midjourney is not owned by Discord; Discord was its original main interface.
  • An attractive image is not automatically accurate, original or legally safe to publish.
  • Prompting is only one part of professional work; selection, editing, typography and rights review remain important.

Timeline: key years and locations

YearLocationEventWhy it mattered
2021San Francisco Bay Area, United StatesMidjourney research laboratory is formedCreated an independent generative-media company.
July 2022Discord and global internet communityPublic beta beginsMade text-to-image generation available to a mass audience.
2022–2023GlobalRapid model versions improve qualityEstablished Midjourney’s distinctive visual reputation.
2023–2024GlobalEditing, references and web tools expandGave creators more control and organization.
2025–2026GlobalImage models and video workflows advanceBroadened the platform from still art toward generative media.

Frequently asked questions

Who founded Midjourney?

David Holz founded Midjourney as an independent research laboratory.

When did Midjourney become public?

Its public beta began in 2022.

Why did Midjourney use Discord?

Discord provided a fast command interface and a built-in community where users could see and learn from generations.

Can Midjourney images be copyrighted?

Copyright treatment depends on jurisdiction, human creative contribution and evolving law; users should review current rules and obtain legal advice for important commercial cases.

Practical lessons for responsible AI image use

Midjourney helped make prompt-based image creation accessible, but a polished image still needs human direction and review. Creators should verify anatomy, text, logos, cultural details and claims before publication. They should also keep records of prompts and edits when client approval or provenance matters. For website use, an image should be resized, compressed and given accurate alternative text instead of being uploaded as an oversized decorative file. The strongest workflow treats AI output as source material: select deliberately, correct errors, establish usage rights and connect the final visual to genuinely useful written content.

Sources and references