AI in Anime Production: The 2024-2025 Debate

The debate over generative AI in anime production has, across 2024 and 2025, hardened from a speculative concern into a concrete industry argument with clearly articulated positions on multiple sides. Studios, animator unions, streaming platforms, and individual creators have all stated public positions; some have changed positions in response to specific incidents.

The argument is structural. Anime production involves many distinct labor stages — storyboards, layouts, key animation, in-betweens, coloring, backgrounds, compositing, subtitle and dub localization — and generative AI tools have very different applicability and acceptance levels at each stage. The 2024-2025 debate has been less about whether AI belongs in anime than about which specific tasks it can or should perform.

The Boy and the Heron, Studio Ghibli’s 2023 Hayao Miyazaki film, has functioned as a touchstone in this conversation — explicitly handmade, with publicly stated rejection of AI tooling, positioned as the artisan-craft counterpoint to industry-wide automation pressure.

The stages where AI has been used

Several specific production stages have seen documented AI tool adoption, with varying levels of public acknowledgment:

In-between frame generation. AI tools that interpolate between two key frames to generate in-between frames have been under exploration at multiple studios. The technical case is straightforward (in-betweens are repetitive labor) but the artistic case is contested (in-betweens contribute to motion quality in ways that automated interpolation can flatten).

Upscaling and resolution conversion. AI-based upscaling tools have been adopted widely for converting older anime (originally produced at standard definition) to high-resolution streaming masters. This use case has been relatively uncontroversial, since it does not replace new creative labor.

Color correction and consistency. AI tools for color-script consistency across episodes and for paint-bucket coloring of line art have been adopted at some studios. The labor displacement here is significant — coloring is a major in-house labor cost — but the technical performance has been mixed.

Subtitle generation. Crunchyroll, Netflix, and other streaming platforms have invested in AI-based subtitle generation tools. The use case is significant: anime simulcasts require fast subtitle turnaround, and AI can produce drafts that human translators then revise. Translator union responses have been mixed.

Dub acceleration. AI voice synthesis and dub-timing tools have been under exploration for localization workflows. Voice actor unions have raised significant objections to potential voice replication uses.

The studios that have stated positions

Several major studios have made public statements about their AI policies, with varying degrees of specificity:

Studio Trigger. Trigger has publicly stated that it does not use AI for character animation. The studio’s house aesthetic — heavily dependent on individual animator personality and sakuga showcases — has been positioned as structurally incompatible with AI automation.

Wit Studio. Wit issued statements during 2024 about limits on AI use in its productions, particularly around character animation. The studio’s prestige-oriented production model has emphasized human-authored animation as a quality signal.

Studio Ghibli. Ghibli’s position has been the most categorical. The Boy and the Heron was explicitly handmade. The studio has not adopted AI tooling in any major way, and Hayao Miyazaki’s well-known 2016 criticism of AI animation (in which he described an AI-generated animation demonstration as “an insult to life itself”) has been widely cited as the studio’s enduring position.

Studio Bind. Studio Bind, producer of Mushoku Tensei, has publicly disclosed limited use of AI tools for some non-character work, primarily background-related tasks. The disclosure was unusual for its specificity; most studios have not detailed their tool use as transparently.

Other prestige studios. Multiple other prestige-tier studios (Ufotable, MAPPA, ufotable, P.A. Works) have made varying statements about limits on AI use without always specifying which tools are used or rejected.

The union and labor position

The Japan Animation Creators Association (JAniCA) and translator/voice-actor unions have generally pushed for explicit guidelines around AI use, focusing on several specific concerns:

Wage displacement. AI tools that displace human labor (in-betweens, coloring, translation) without corresponding wage protections worsen an already-precarious labor market.

Attribution and consent. AI training data that includes existing anime work raises questions about whether animators and studios have consented to having their work used as training material.

Voice and likeness rights. Voice actor unions have particularly emphasized the need for explicit consent and compensation for any AI voice replication uses.

Transparency. Unions have called for studios and platforms to disclose AI use to audiences and to other production participants.

Industry-wide guidelines on these questions have not been formalized as of 2026. Individual studios and platforms have made varying disclosures.

The streaming platform investment

Crunchyroll and Netflix have both invested substantially in AI tooling for localization work — subtitle generation, dub workflow acceleration, audience-language expansion. The platforms’ position has generally been that AI tools enable broader and faster localization (reaching more audiences in more languages) while still requiring human translators and voice actors for final quality.

The translator-side response has been mixed. Some translators have welcomed AI tools as drafting assistance; others have raised concerns about wage compression and quality degradation. The professional translator community is still working out its collective position.

The 2024 statements moment

A particular cluster of statements during 2024 — from Wit Studio, Studio Trigger, and several others — clarified that a portion of the prestige-tier studio community had decided to publicly differentiate themselves from AI-adopting workflows. This was strategically significant: in a market where audiences increasingly track studio choices, publicly rejecting AI for character animation became a brand position.

The same year, several streaming platforms expanded their AI tooling deployments and made public statements defending the use cases (typically: localization speed, accessibility expansion, not replacement of creative labor). The argument map became clear: studios on one side defending human-craft authorship of character animation; platforms on another side defending AI’s role in localization and access; and animator unions advocating for explicit protections across both.

The Miyazaki position and the Heron’s framing

Hayao Miyazaki’s 2016 commentary on an AI animation demonstration — in which he described the demonstration as fundamentally disrespectful to the artistic tradition he had spent his life building — became a widely-cited reference point in the 2024-2025 debate. Whether he intended it that way or not, the comment functioned as a position statement that aligned naturally with Ghibli’s continued commitment to handmade production.

The Boy and the Heron’s 2023 release and 2024 Academy Award for Best Animated Feature gave additional weight to this position. The film’s commercial and critical success, on explicitly handmade terms, made the artisan-craft argument structurally available to other studios that wanted to align with it.

What 2026 looks like

By mid-2026, the industry has reached partial equilibrium without consensus. Localization-related AI use is widely deployed across streaming platforms. Coloring and in-between AI tools are used at some studios and rejected at others. Character animation by AI remains commercially marginal and reputationally compromised.

What changes the picture further is likely to depend on three things: the quality of AI tools as they continue to develop, the formation of explicit industry guidelines (which JAniCA and others continue to push for), and audience response to disclosed AI use. None of these has been fully resolved by 2026. The debate continues.