David Holz
Midjourney v4 release announcement (November 2022)
Midjourney v4 era painterly aesthetic. Highly stylized digital-paint look, rim-lit dramatic lighting, ornamental detail, characteristic 2023 default polish.
The moves a human crew would use to get here. Cue applies them for you, and knowing their names is how you ask for a change.
use a three-point Rembrandt-style lighting reference in prompts to activate the subsurface oil-portrait mode
request 85mm f/1.4 or 'shallow depth of field' to trigger the characteristic lens simulation
add slight warm shadow tint and pull highlights toward neutral gray in post
apply Orton effect at 15% Screen opacity to boost the soft-glow quality of v4 midtones
add subtle canvas grain at 20-30 units to reinforce the painterly material quality
weighted composition crop: v4 compositions often center the subject; respecting this in final crop maintains the characteristic look
cool split: v4 defaults toward orange-teal contrast - enhance with Color Grading wheel in post
Midjourney v4 was released in November 2022 by Midjourney Inc., the San Francisco AI research lab founded by David Holz (previously co-founder of Leap Motion). The release represented a step-change from earlier Midjourney models: v4 produced dramatically more coherent human anatomy (including the first model version that could render hands with acceptable finger counts reliably), richer material texture simulation, and a characteristic aesthetic blend that sits between photorealism and oil painting. This blend - the specific painterly-but-photographically-lit quality that the model consistently defaulted to without explicit style prompting - became the defining AI image aesthetic of early-to-mid 2023.
V4 was trained on a substantially larger and higher-quality dataset than its predecessors, with particular attention to fine art, illustration, and photography. The model learned to synthesize coherent depth cues, volumetric lighting, and material behavior simultaneously. The resulting images typically show:
Midjourney v4's images proliferated across social media in late 2022 and through 2023, becoming the visual signature of the "AI image" era in mainstream culture. They appeared in marketing campaigns (when not disclosed), book covers (Zarya of the Dawn copyright controversy, 2022-2023), editorial illustrations, and personal creative projects at an unprecedented scale. The specific look became so recognizable that it was both widely adopted and widely satirized.
V5 (released March 2023) shifted toward higher photorealism and reduced the painterly quality; v6 (late 2023) added more natural composition variability. V4's painterly mid-point aesthetic became retrospectively identified as the "AI art look" of 2023.
Midjourney v4 release announcement (November 2022)
proliferation in social media, editorial design, marketing (2022-2023)
first AI-assisted comic with Midjourney v4 imagery and US copyright case (2022-2023)
won Colorado State Fair digital art competition using Midjourney v4 (September 2022)
2023
_PSEUDOAMNESIA: The Electrician_ AI-generated photograph submitted to and declined at Sony World Photography Awards
A Look is not a filter. It is a set of stored decisions the pipeline reads before it makes a single frame, so two videos in the same Look actually match.