79% of music makers now use AI for technical work such as mixing, mastering or audio restoration, according to LANDR’s 2025 study of more than 1,200 creators, making mastering the most normalized entry point for AI in music production. The demand side explains why: streaming services received an average of 106,000 new tracks per day in 2025, and 96.2% of them came from independent and DIY distribution rather than major labels (Luminate 2025 Year-End Music Report). At the same time, only 32% of producers in Tracklib’s 1,734-person 2025 study use any AI tool at all, and human engineers still won the top two spots in a 472-listener blind mastering test. We aggregated data from LANDR, Tracklib, Ditto Music, Luminate, Spotify, the Audio Engineering Society (AES), iZotope, Market.us, Grand View Research, and Future Market Report. For the human side of the business, see our mastering industry statistics.
TL;DR
- 87% of music makers use AI somewhere in their workflow, and 79% use it for mixing, mastering or restoration (LANDR AI in Music Study 2025).
- Only 32% of producers use any AI tool, and just 6% use fully generative tools regularly (Tracklib Music Producers and AI 2025).
- 45.5% of AI-using producers apply it to mastering and EQ plugins, second only to stem separation at 73.9% (Tracklib 2024).
- 66% of independent artists said they would use AI to mix and master their music (Ditto Music 2023).
- LANDR has mastered more than 25 million songs for over 5 million artists in a decade (LANDR, October 2023).
- 106,000 new tracks reached streaming services per day in 2025, up 7% from 99,000 in 2024 (Luminate 2025 Year-End Report).
- Spotify normalizes to -14 LUFS by default and asks for true peak below -1 dBTP (Spotify for Artists).
- The AES recommends -16 LUFS distribution loudness for music and -18 LUFS for speech (AES TD1008, 2021).
- The AI in music market is projected to grow from USD 5.20 billion in 2024 to USD 60.44 billion by 2034 (Market.us).
- The mastering software market is sized at USD 985.75 million in 2025 (Future Market Report 2025).
- 82.2% of producers who avoid AI cite artistic reasons, and 34.5% cite quality (Tracklib 2024).
- Human engineers placed 1st and 2nd against AI tools in a 472-listener blind mastering test (Benn Jordan study, 2024).
1. AI Adoption Among Music Makers: The Headline Numbers
The first thing to understand about AI adoption in music production is that the headline rate depends heavily on who you ask. Survey results range from 25% to 87%, and the gap is mostly a sampling story, not a contradiction. Platforms that sell AI tools survey their own users; neutral sample marketplaces and distributors reach a broader, more skeptical base.
MusicTech’s analysis of the 2026 Water and Music and Moises study showed the effect clearly: 80% of its 1,525 respondents were Moises customers, and it reported that 78% of professional musicians use AI. The same article notes that LANDR’s 87% figure came from 1,241 respondents drawn from LANDR’s own users. Read these numbers as “adoption among platform-connected creators”, not as a census of every producer.
| Metric | Value | Source |
|---|---|---|
| Music makers using AI in at least one part of their process | 87% | LANDR AI in Music Study 2025 |
| Respondents using more AI tools than the previous year | 69% | LANDR AI in Music Study 2025 |
| Producers using any AI tool | 32% | Tracklib Music Producers and AI 2025 |
| Producers using fully generative AI tools regularly | 6% | Tracklib Music Producers and AI 2025 |
| Producers using AI (prior year) | 25% | Tracklib AI Music Production Survey 2024 |
| Independent artists who used AI in music-making | 48% (2025) vs 59.5% (2023) | Ditto Music AI survey 2025 |
| Professional musicians using AI (Moises-heavy sample) | 78% | Water and Music / Moises study 2026 |
Context note: Tracklib’s two waves are the closest thing to a like-for-like trend line. Producer AI use rose from 25% (1,107 respondents, June 2024) to 32% (1,734 respondents, November 2025), per MusicTech’s coverage of the 2025 study. Ditto Music’s 2025 survey moved the other way, from 59.5% to 48%, which suggests adoption is plateauing outside the most AI-friendly user bases.
2. Mastering and Technical Tasks: Where AI Lands First
Across every survey that breaks out use cases, assistive and technical tasks lead and generative songwriting trails. 79% of LANDR’s respondents use AI for mixing, mastering or audio restoration, compared with 66% for creative tasks and 52% for visual and promotional work. Mastering is the rare AI use case that most producers frame as a tool, not a threat to authorship.
Tracklib’s 2024 survey, reported by Music Business Worldwide, puts mastering in second place among producers who already use AI: 45.5% use AI mastering and EQ plugins, behind stem separation at 73.9% (the subject of our AI audio separator statistics). Fully generated songs sat at 3%. The question for mastering is no longer acceptance; it is whether people will pay, since nearly 73% of AI users relied only on free tools.
| Metric | Value | Source |
|---|---|---|
| Use AI for technical tasks (mixing, mastering, restoration) | 79% | LANDR AI in Music Study 2025 |
| Use AI for creative tasks (songwriting, melodies, instruments, vocals) | 66% | LANDR AI in Music Study 2025 |
| Use AI for visual and promotional work | 52% | LANDR AI in Music Study 2025 |
| AI-using producers applying it to mastering and EQ plugins | 45.5% | Tracklib AI Music Production Survey 2024 |
| AI-using producers applying it to stem separation | 73.9% | Tracklib AI Music Production Survey 2024 |
| AI-using producers generating entire songs | 3% | Tracklib AI Music Production Survey 2024 |
| AI users relying solely on free tools | Nearly 73% | Tracklib AI Music Production Survey 2024 |
| Artists who would use AI to mix and master their music | 66% | Ditto Music survey 2023 |
Outlier note: Ditto Music’s 2023 survey of 1,200+ users measured intent, not usage. Mixing and mastering (66%) ranked above songwriting (47%) but below album artwork (77%) and production (62%). Most recent available data for that question: Ditto Music, 2023.
3. Platform Scale: LANDR, BandLab and the Automated Mastering Base
Disclosed platform numbers show automated mastering became a mass-market service years before generative music did. LANDR reported more than 25 million songs mastered for over 5 million artists across a decade when it launched its AI mastering plugin for DAWs in October 2023, per its press release. Two years later its study announcement cited more than 7 million creators on the platform.
BandLab, which bundles free mastering inside a browser and mobile DAW, crossed 100 million registered users in March 2024, according to Music Business Worldwide. When mastering ships as a default button inside the DAW, adoption is measured in installed base, not purchase decisions. That shift also pressures the paid plug-in tier tracked in our audio plug-in market statistics.
| Metric | Value | Source |
|---|---|---|
| Songs mastered by LANDR’s AI in its first decade | 25 million+ | LANDR press release, October 2023 |
| Artists who had used LANDR mastering | 5 million+ | LANDR press release, October 2023 |
| Years of AI mastering research behind LANDR’s plugin | 10 | LANDR press release, October 2023 |
| Creators on the LANDR platform | 7 million+ | LANDR study announcement, November 2025 |
| BandLab registered users | 100 million (March 2024) | Music Business Worldwide, 2024 |
| BandLab registered users one year earlier | 60 million (January 2023) | Music Business Worldwide, 2024 |
| Share of BandLab users based in the US | 30% | Music Business Worldwide, 2024 |
| Growth in BandLab music creation using AI tools | 15x (as of August 2024) | Music Business Worldwide, 2024 |
Context note: these are company-disclosed figures without third-party audit. LANDR’s 25 million figure is cumulative and dates from 2023; no newer mastered-track total has been published.
4. The Demand Side: Upload Volume That Needs Mastering
Automated mastering exists because the volume of new releases has outgrown the human mastering workforce. 106,000 new tracks were delivered to streaming services every day in 2025, up 7% from 99,000 in 2024, per Luminate’s year-end data reported by Music Business Worldwide. Independent and DIY distributors accounted for 96.2% of those deliveries.
That mix matters for mastering economics. Indie releases rarely carry the budget for a booked engineer per track, and 21.9% of all tracks on streaming services earned zero streams in 2025. For a release with uncertain upside, a fast, low-cost automated master is the rational default. By September 2026, Luminate CEO Rob Jonas put the daily figure at close to 150,000 tracks, per Music Ally.
| Metric | Value | Source |
|---|---|---|
| New tracks (ISRCs) delivered to streaming services per day | 106,000 (2025) | Luminate 2025 Year-End Music Report |
| Daily deliveries one year earlier | 99,000 (2024) | Luminate 2025 Year-End Music Report |
| Total tracks on audio streaming services | 253 million (end of 2025) | Luminate 2025 Year-End Music Report |
| Share of daily deliveries from independent and DIY distribution | 96.2% | Luminate 2025 Year-End Music Report |
| Share of deliveries from major-label distribution | 3.8% (down from 8% in 2024) | Luminate 2025 Year-End Music Report |
| Tracks with zero streams in 2025 | 55.3 million (21.9%) | Luminate 2025 Year-End Music Report |
| Tracks uploaded per day, September 2026 | Close to 150,000 | Luminate CEO Rob Jonas, via Music Ally 2026 |
| Fully AI-generated tracks delivered to Deezer per day | Nearly 90,000 (June 2026) | Deezer, via Music Business Worldwide 2026 |
Outlier note: part of the volume growth is fully AI-generated music, which arrives already rendered and often already loudness-processed. Deezer said fully AI-generated tracks passed 50% of new deliveries on peak days in June 2026, up from 75,000 a day (44%) in April, and that it tagged more than 13.4 million AI tracks across 2025, per Music Business Worldwide. Those tracks still drew only 1% to 3% of streams.
5. Streaming Loudness Targets AI Masters Must Hit
Most AI mastering engines expose a loudness target, and the reason is normalization: streaming services turn loud masters down to a reference level, so excess limiting buys nothing but distortion. Spotify normalizes to -14 LUFS integrated by default and advises keeping true peak below -1 dBTP, or below -2 dBTP for masters louder than -14 LUFS, according to its loudness normalization guide.
The targets are not uniform. iZotope’s July 2025 platform guide lists Apple Music at -16 LUFS, Deezer at -15 LUFS and SoundCloud with no normalization at all. The standards body sits lower than most services: the AES TD1008 recommendation sets -16 LUFS distribution loudness for music and -18 LUFS for speech, with true peak no higher than -1 dBTP. The practical takeaway is that a single “streaming preset” is a compromise, which is why most automated masters land somewhere between these numbers.
| Metric | Value | Source |
|---|---|---|
| Spotify default normalization target | -14 LUFS integrated | Spotify for Artists, Loudness normalization |
| Spotify Premium ‘Loud’ and ‘Quiet’ settings | -11 LUFS and -19 LUFS | Spotify for Artists, Loudness normalization |
| Spotify true peak guidance | Below -1 dBTP (below -2 dBTP if louder than -14 LUFS) | Spotify for Artists, Loudness normalization |
| AES recommended distribution loudness, music | -16 LUFS | AES TD1008, 2021 |
| AES recommended distribution loudness, speech | -18 LUFS | AES TD1008, 2021 |
| Apple Music normalization target | -16 LUFS | iZotope, Mastering for streaming platforms 2025 |
| YouTube, Amazon Music and Tidal targets | -14 LUFS | iZotope, Mastering for streaming platforms 2025 |
| Deezer target / SoundCloud | -15 LUFS / no normalization | iZotope, Mastering for streaming platforms 2025 |
Context note: iZotope suggests a practical master around -12 LUFS integrated with peaks near -1 dBTP, which every normalizing service above would turn down slightly. Most recent available data for the standards row: AES TD1008, 2021.
6. Market Size and Growth Forecasts
There is no clean market estimate for AI mastering alone; analysts fold it into either “AI in music” or “mastering software”. The resulting figures span more than an order of magnitude. Market.us sizes the AI in music market at USD 5.20 billion in 2024 and projects USD 60.44 billion by 2034, a 27.8% CAGR, with music production and recording at 33.9% of 2024 revenue, per its report page.
Narrower definitions give smaller numbers. Grand View Research puts generative AI in music at USD 440.0 million in 2023, reaching USD 2.79 billion by 2030 (report summary). Future Market Report sizes all mastering software, AI and conventional, at USD 985.75 million in 2025, growing at 9.16% a year. The spread says more about scope than about growth: AI mastering is a feature inside larger categories, which is also how it shows up in our AI music generation statistics.
| Metric | Value | Source |
|---|---|---|
| AI in music market, 2024 | USD 5.20 billion | Market.us, AI in Music Market report |
| AI in music market forecast, 2034 | USD 60.44 billion (27.8% CAGR, 2025-2034) | Market.us, AI in Music Market report |
| Music production and recording share of AI in music, 2024 | 33.9% | Market.us, AI in Music Market report |
| Software and cloud shares of AI in music, 2024 | 64.7% software, 73.5% cloud deployment | Market.us, AI in Music Market report |
| Generative AI in music market | USD 440.0 million (2023) to USD 2.79 billion (2030), 30.4% CAGR | Grand View Research, Generative AI in Music Market report |
| Mastering software market | USD 985.75 million (2025) to USD 1,987.40 million (2033), 9.16% CAGR | Future Market Report, Mastering Software Market, August 2025 |
| Music production share of mastering software revenue | 46.7% | Future Market Report, Mastering Software Market, August 2025 |
| North America share of mastering software market | 35.4% | Future Market Report, Mastering Software Market, August 2025 |
Conflict note: Market.us itself has published two editions with different numbers. Its January 2025 release put the 2024 market at USD 4.9 billion and the 2033 forecast at USD 38.7 billion (25.8% CAGR); the current page shows USD 5.20 billion and USD 60.44 billion. Treat all forecasts as directional.
7. Trust, Quality and Resistance
Adoption of AI mastering has outpaced trust in AI music overall. 38% of producers are actively against AI and only 17% actively favor it, with 45% neutral, in Tracklib’s 2025 study. Among the 75% of producers not using AI in 2024, 82.2% cited artistic reasons and 34.5% said the quality was not good enough, which is the objection automated mastering has to answer directly.
The listening evidence is mixed rather than damning. A 2022 Journal of the AES paper by Mitchell Elliott and Song Hui Chon (Belmont University) compared two human engineers with two automatic mastering services and found listeners could not consistently pick the human masters; they preferred human work on jazz but not on rock. In a larger, less formal blind test run by producer Benn Jordan in 2024, 472 listeners rated seven finalists and human engineers took first and second place, while LANDR was eliminated before the final round, per MusicTech. The pattern is consistent: automated masters are good enough to be hard to spot, and not yet good enough to win when experts compete. For the studio side of that comparison, see our music mastering studio statistics.
| Metric | Value | Source |
|---|---|---|
| Producers actively for / neutral / against AI | 17% / 45% / 38% | Tracklib Music Producers and AI 2025 |
| Non-users citing artistic or creative reasons | 82.2% | Tracklib AI Music Production Survey 2024 |
| Non-users citing insufficient quality | 34.5% | Tracklib AI Music Production Survey 2024 |
| Non-users citing cost / copyright | 14.3% / 10.2% | Tracklib AI Music Production Survey 2024 |
| Artists naming lack of personal creativity as top AI concern | 42% (2025) vs 23% (2023) | Ditto Music AI survey 2025 |
| Artists who say they would never use AI | 18% (2025) vs 29% (2023) | Ditto Music AI survey 2025 |
| Music makers expecting AI to hurt / help / both | 43.6% / 24.8% / 31.6% | LANDR AI in Music Study 2025 |
| Listeners in largest public blind mastering test | 472 (human engineers placed 1st and 2nd) | Benn Jordan mastering study, 2024 |
Context note: the Benn Jordan test used one track and a self-selected online audience, so it is closer to a large field experiment than a peer-reviewed study. The AES paper used a small, controlled design. Neither should be read as a verdict on every engine or genre. Most recent available peer-reviewed data: Journal of the AES, 2022.
Summary: AI Audio Mastering Adoption by the Numbers
| Metric | Value | Source |
|---|---|---|
| Music makers using AI somewhere in their workflow | 87% | LANDR AI in Music Study 2025 |
| Music makers using AI for mixing, mastering or restoration | 79% | LANDR AI in Music Study 2025 |
| Producers using any AI tool | 32% | Tracklib Music Producers and AI 2025 |
| Producers using fully generative tools regularly | 6% | Tracklib Music Producers and AI 2025 |
| AI-using producers applying it to mastering and EQ | 45.5% | Tracklib AI Music Production Survey 2024 |
| AI users relying solely on free tools | Nearly 73% | Tracklib AI Music Production Survey 2024 |
| Artists who would use AI to mix and master | 66% | Ditto Music survey 2023 |
| Independent artists who used AI | 48% (2025), down from 59.5% (2023) | Ditto Music AI survey 2025 |
| Songs mastered by LANDR in its first decade | 25 million+ | LANDR press release, October 2023 |
| BandLab registered users | 100 million (March 2024) | Music Business Worldwide, 2024 |
| New tracks delivered to streaming services per day | 106,000 (2025) | Luminate 2025 Year-End Music Report |
| Share of deliveries from independent and DIY distribution | 96.2% | Luminate 2025 Year-End Music Report |
| Spotify default normalization target | -14 LUFS | Spotify for Artists |
| AES recommended distribution loudness for music | -16 LUFS | AES TD1008, 2021 |
| Apple Music normalization target | -16 LUFS | iZotope, Mastering for streaming platforms 2025 |
| AI in music market forecast | USD 60.44 billion by 2034 | Market.us |
| Mastering software market | USD 985.75 million (2025) | Future Market Report 2025 |
| Producers actively against AI | 38% | Tracklib Music Producers and AI 2025 |
| Non-users citing insufficient quality | 34.5% | Tracklib AI Music Production Survey 2024 |
| Listeners in largest public blind mastering test | 472 | Benn Jordan mastering study, 2024 |
Methodology and Sources
- LANDR, AI in Music Study 2025 (1,200+ respondents, fielded September 30 to October 6, 2025): landr.com/ai
- LANDR, AI mastering plugin press release, October 2023: landr.com
- Tracklib, AI Music Production Survey 2024 (1,107 producers), via Music Business Worldwide
- Tracklib, Music Producers and AI 2025 (1,734 producers), via MusicTech
- Water and Music / Moises study 2026 and survey sampling analysis, via MusicTech
- Ditto Music, AI survey 2023: press.dittomusic.com
- Ditto Music, AI survey 2025: press.dittomusic.com
- BandLab user milestone, via Music Business Worldwide
- Luminate, 2025 Year-End Music Report, via Music Business Worldwide
- Luminate CEO daily upload figure, September 2026, via Music Ally
- Deezer AI upload data, 2026, via Music Business Worldwide
- Spotify for Artists, Loudness normalization: support.spotify.com
- Audio Engineering Society, TD1008 Recommendations for Loudness of Internet Audio Streaming and On-Demand Distribution (2021): aes.org
- iZotope, How to master for streaming platforms (July 2025): izotope.com
- Market.us, AI in Music Market report: market.us; earlier edition (January 2025) via EIN Presswire
- Grand View Research, Generative AI in Music Market report: grandviewresearch.com
- Future Market Report, Mastering Software Market (August 2025): futuremarketreport.com
- Elliott and Chon, A Comparative Study of Music Mastered by Human Engineers and Automated Services, Journal of the AES vol. 70 no. 9 (2022): aes.org
- Benn Jordan blind mastering study (2024), via MusicTech
- Data watch: Adoption surveys are not comparable one-to-one. LANDR and the Water and Music / Moises study recruited from their own user bases, which inflates AI usage; Tracklib and Ditto reach broader samples. Market sizing conflicts are large: Market.us (USD 5.20 billion, 2024), Grand View Research (USD 440.0 million generative AI in music, 2023) and Future Market Report (USD 985.75 million mastering software, 2025) measure different scopes, and Market.us revised its own forecast between editions. A widely repeated claim that 30.6% of artists use AI for mastering, attributed to Ditto Music, could not be found on Ditto’s published survey pages and was excluded. LANDR’s mastered-track total dates from 2023, the AES loudness guideline from 2021, and the peer-reviewed human vs AI mastering comparison cited here from 2022; Grand View Research figures are taken from its public report summary. Newer Tracklib and Ditto survey waves are expected in 2026.
Last updated: October 3, 2026. We update this roundup quarterly, and the next refresh is expected when Luminate publishes its 2026 Year-End Music Report.