Who Is Already Sharing the Charts with Algorithms?
AI Music in the Global Market: Where to Measure It, How to Compare It with Human-Led Releases, and How Far It Has Come by August 2026
The music industry has reached a point where the question, “Can music created with the help of artificial intelligence enter the charts?” is already outdated.
The answer is yes — it can, and it already has.
AI-related releases have reached No. 1 on individual Billboard charts, topped national Spotify Top 50 rankings, entered the official German Top 50, climbed to the upper positions of UK sales and download charts, and, by the summer of 2026, a track surrounded by serious discussion over the use of generative AI had reached the Billboard Hot 100 — one of the world’s most important mainstream music rankings.
Yet there is a paradox:
we can measure the popularity of a song far more precisely than we can measure how it was created.
Billboard, Spotify, Apple Music, YouTube, Shazam, national charts, Luminate, Soundcharts and Chartmetric are all very good at showing us what people are listening to.
Until recently, however, almost none of these sources systematically answered another question:
How much of a particular recording was created by humans, and how much was created with generative systems?
That is why today there is still no single authoritative global ranking with columns such as:
Human — AI-Assisted — AI-Generated.
But in 2026 the situation began to change extremely quickly.
Industry definitions for AI-Generated and AI-Assisted appeared. AI metadata is emerging. Automated detection systems are becoming available. AI-specific charts are being created.
And, perhaps most importantly, official music industries are beginning to establish rules governing whether and how AI-involved recordings are eligible for national charts.
What we are witnessing, therefore, is not simply the rise of a new type of content.
A new system for measuring it is being built in real time.
Open the questions one by one.
Each chapter can be opened and folded again. The full text and source links are preserved.
011. First, We Need to Define What Exactly We Are Comparing+
One of the most common methodological mistakes in discussions about AI music is to divide all music into two categories:
“made by a human” / “made by AI.”
Modern production no longer works that way.
In one recording, a human may write the lyrics and melody, play the guitar and record the vocal, while a generative system creates selected instrumental parts.
In another, AI may generate the music and vocals, while a human edits the structure and writes the lyrics.
In a third, nearly the entire piece may emerge from a prompt, with the creator mainly selecting the result.
On July 10, 2026, a coalition of music industry organizations including IFPI, RIAA, A2IM, WIN, IMPALA and others introduced two important categories.
AI-Generated — generative AI has created the entire work or a substantial proportion of its creative elements, such as the lead vocal, principal instrumental performances, or most of the music from a prompt.
AI-Assisted — the work remains substantially human-created; the lead vocal and primary instrumental performances are human, while GenAI is used for some expressive or production elements.
This distinction is fundamental.
For the purposes of serious market research, it therefore makes more sense to use at least three categories:
| CategoryMeaning | |
|---|---|
| Human-led / no disclosed GenAI | Human artist and production process in which significant use of GenAI has not been publicly disclosed |
| AI-Assisted / Human + AI | Core authorship and performance remain human, while AI is part of the creative or production process |
| Primarily / Fully AI-Generated | Vocals, instrumentals, or a substantial part of the musical material are generated by AI |
The first category is deliberately called “no disclosed GenAI”, rather than “100% human.”
The absence of information about AI is not proof that AI was completely absent.
This is the first major problem in any statistical comparison of the market.
022. How Large Is the Market in Which This Experiment Is Taking Place?+
According to IFPI, the global recorded music industry generated approximately $31.7 billion in 2025, representing growth of 6.4% and marking the eleventh consecutive year of expansion.
Streaming generated more than $22 billion and accounted for approximately 69.6% of global recorded music revenues.
The number of paid streaming subscription users reached approximately 837 million.
So when we discuss the penetration of AI into music, we are no longer talking about a small experimental internet culture.
We are talking about a technology beginning to interact with an enormous existing system that distributes attention, visibility and money.
And here we encounter one of the most striking figures in the entire debate.
More Than Half of New Music — But Only a Few Percent of Listening
On July 21, 2026, Deezer reported that at the June peak, more than 50% of all tracks delivered to the platform each day were identified as fully AI-generated.
That represented approximately 90,000 AI tracks per day.
Yet those recordings accounted for only about 1–3% of actual streams on Deezer.
This may be one of the most important statistics for understanding the market today.
The quantity of content being produced and the quantity of human attention are entirely different metrics.
AI is already capable of dominating supply.
It has not yet dominated demand.
Deezer has also reported that during 2025 as much as 85% of streams involving fully AI-generated content identified by the platform were associated with fraudulent activity and therefore excluded from royalty calculations.
For Deezer’s catalogue overall, the corresponding level of fraudulent streams was substantially lower, at roughly 8%.
These figures belong specifically to Deezer and should not automatically be generalized to Spotify, Apple Music or the global market as a whole.
Nevertheless, as an indicator they are extremely significant.
So the statement:
“Half of all new music is already being created by AI”
may be technically true for the upload flow of a particular platform while simultaneously creating a completely misleading impression of what people are actually listening to.
033. What Is Actually at the Top of the Mainstream Global Market?+
For comparison, we can look at IFPI’s global ranking.
In the 2025 Global Single Chart, ROSÉ & Bruno Mars — “APT.” ranked No. 1 with 2.06 billion Global Subscription Stream Equivalents.
It was followed by tracks including “Golden,” “Ordinary” and other major global releases.
The upper reaches of the largest global rankings therefore remain dominated by projects centered around known human artists.
But methodological caution is necessary.
From such a ranking we cannot conclude:
“All of these recordings were made without AI.”
What we can say is:
“These recordings are not presented to the market as AI-generated projects.”
Before the emergence of standardized or mandatory AI metadata, the industry generally did not collect enough information to give a more precise answer.
That is why a traditional Top 100 and a future AI Transparency Top 100 would be two very different things.
044. Where Confirmed AI Hits Have Already Appeared+
The history of AI music entering the charts has not developed in one straight line.
Early successes appeared in platform charts, viral rankings, download charts and specialist genre charts.
Only later did AI begin to approach broader national rankings.
| ReleaseMarket / ChartResultWhy It Matters | |||
|---|---|---|---|
| Butterbro feat. udio.com — “Verknallt in einen Talahon” | Germany | #48 German Top 50 | An early example of an AI-generated composition entering a major national all-market chart |
| Xania Monet | Billboard, US | #1 R&B Digital Song Sales, #3 Hot Gospel Songs, #30 Adult R&B Airplay | AI-related project entered not only sales charts but radio airplay |
| Breaking Rust — “Walk My Walk” | Billboard Country Digital Song Sales | #1 | Major AI milestone, but in a narrow paid-download chart rather than the Billboard Hot 100 |
| Jacub — “Jag vet, du är inte min” | Spotify Sweden Top 50 | #1 | AI hit became the most popular track on Spotify in Sweden |
| The same Jacub track | Official Sverigetopplistan | Not admitted | The same song could be No. 1 on Spotify while having no official national chart position |
| Eddie Dalton — “Another Day Old” | UK Official Singles Sales | #2 | AI persona reached the upper tier of an official UK sales ranking |
| Fenix Flexin — “RUBBERZ” | Billboard Hot 100 | #58 | A major 2026 threshold: an AI-involved track appeared in the broad all-genre Hot 100 |
Let us examine several of these cases more closely.
055. Germany: AI Moves from Viral Charts into a National Top 50+
In August 2024, “Verknallt in einen Talahon” by Butterbro entered the German Top 50 at No. 48.
Creator Josua Waghubinger used Udio to generate vocals and instrumental material.
By the time the song entered the ranking, it had accumulated approximately 3.5 million Spotify streams and had reached No. 3 on the Spotify Global Viral Chart.
This case matters because AI had moved beyond being a technological demonstration.
People were not simply clicking on an AI song because it was novel or strange.
Consumption had become large enough for the track to enter a real national ranking in one of the world’s largest music markets.
But No. 48 is not chart domination.
It is evidence of market entry.
066. Xania Monet: AI Enters Billboard — and Radio+
Xania Monet became one of the most important American AI-related music projects of 2025.
Behind the project is Telisha Jones, who writes the lyrics, while the musical output is created using Suno.
Billboard recorded several notable results for the project.
“How Was I Supposed to Know?” entered Adult R&B Airplay, debuting at No. 30.
The project also appeared on Hot R&B Songs, Hot Gospel Songs and Emerging Artists, while one track reached No. 1 on R&B Digital Song Sales.
The radio airplay result is particularly significant.
Digital sales can sometimes be concentrated among a relatively small number of highly motivated buyers.
Radio airplay represents a different form of institutional penetration: radio stations are actively programming the song.
In other words, AI-related music has begun passing through traditional music-industry filters, not only streaming algorithms.
077. Breaking Rust: Why “AI Reached No. 1 on Billboard” Can Be Misleading+
In November 2025, the AI-generated country project Breaking Rust reached No. 1 on Billboard Country Digital Song Sales with “Walk My Walk.”
Associated Press investigated both the project itself and questions surrounding similarities between its sound and that of Blanco Brown.
But an extremely important qualification is necessary.
This was not the Billboard Hot 100.
Nor was it the broader Hot Country Songs chart, which combines streaming, radio airplay and sales.
Country Digital Song Sales measures a much narrower segment of the market: purchases of digital tracks.
Therefore these two statements:
“An AI song reached No. 1 on Billboard”
and
“An AI song reached No. 1 on Billboard Country Digital Song Sales”
can technically describe the same event while creating completely different impressions of scale.
Serious research into AI music must therefore always specify the exact chart.
088. Sweden: One Song, Two Opposite Realities+
In January 2026, perhaps the clearest methodological example appeared.
Jacub — “Jag vet, du är inte min” reached No. 1 on Spotify Top 50 Sweden.
The creators confirmed that Jacub’s voice was entirely AI-generated and that AI services and Logic had been used in production.
The song accumulated approximately five million Spotify streams.
For Spotify users, this was the biggest track in the country.
But the official Swedish chart, Sverigetopplistan, refused to include it because the recording was judged to be predominantly AI-generated.
The result is remarkable:
Spotify: No. 1.
Official national chart: no position at all.
This is not a statistical error.
The two charts are simply answering different questions.
Spotify asks:
“What are Spotify users listening to right now?”
The industry chart asks:
“Which recordings, according to our rules, are eligible to participate in the official music chart?”
With the rise of AI, this distinction becomes fundamental.
099. United Kingdom: Eddie Dalton and the Power of Download/Sales Charts+
In 2026, the AI-generated soul persona Eddie Dalton also demonstrated why the type of chart matters.
According to the Official Charts Company, “Another Day Old” reached No. 2 on the Official Singles Sales Chart and No. 2 on the Official Singles Downloads Chart.
Other recordings from the project also appeared in sales and download rankings.
This is a meaningful commercial result.
But once again, it cannot automatically be translated into the statement:
“the second most popular song in the United Kingdom.”
The main UK Official Singles Chart measures a much broader range of consumption.
Sales and download charts remain important, but they represent a specific segment of the market.
This once again demonstrates the need to classify not only the recordings themselves, but also the strength and type of evidence provided by each chart.
1010. “RUBBERZ”: AI Reaches the Billboard Hot 100+
By the summer of 2026, a much more significant case had appeared.
“RUBBERZ” by Fenix Flexin reached No. 58 on the Billboard Hot 100.
This is no longer a specialist download chart.
It is the main American all-genre ranking.
The origin of the track became the subject of controversy.
Fenix initially denied AI generation.
However, Treblo — the company whose generator was suspected of being used — ran the recording through its own classifier.
The company reported the song as “very likely Treblo” with a high confidence level and stated that its detector had a false-positive rate below 1 in 10,000.
Fenix later stopped categorically denying the use of AI, although he continued to reject the idea that AI had replaced the recording process itself.
For methodological purposes, it is therefore safer to describe “RUBBERZ” not as an undisputed “first fully AI-generated song on the Hot 100,” but as:
a major Billboard Hot 100 case involving substantial technical and public evidence of significant generative AI use.
Its historical significance remains considerable.
AI-involved music had moved beyond viral, genre and download rankings and entered one of the central showcases of mainstream American music.
1111. Where Can This Information Actually Be Found Today?+
The problem is not that data does not exist.
The problem is that the data lives in different systems that answer different questions.
| SourceWhat It ShowsDoes It Identify AI?Best Use | |||
|---|---|---|---|
| Billboard / Luminate | US charts: streams, sales, radio and combinations of these | Partly; AI identification is developing | Core source for the US market |
| IFPI | Global market, global charts, national chart infrastructure | Now developing formal AI rules | International comparison |
| Official Charts Company | UK charts | AI may appear, but chart type is crucial | UK market |
| Spotify Charts | Actual Spotify consumption by territory | Not a universal AI detector | Fast-moving platform hits |
| Apple Music / iTunes | Streaming rankings and digital purchases | AI metadata is emerging | Important supplementary signal |
| YouTube / TikTok / Shazam | Video, virality, discovery and recognition | No universal AI classification | Early trend discovery |
| Deezer | Streaming plus proprietary AI detection | Yes | One of the strongest current sources for AI analytics |
| Soundcharts | Aggregation of DSP charts, playlists, radio and social data | AI classification is not its main function | Comparing performance across markets |
| Chartmetric | Cross-platform artist and track analytics | No universal AI provenance layer | Artist and track momentum |
| SIQA | Dedicated AI music rankings | Yes, specifically | Monitoring the AI-native market |
Billboard / Luminate
Luminate is one of the core data providers of the US music industry and a long-standing partner of Billboard.
Its databases combine hundreds of verified sources across streaming, retail and airplay.
For serious monitoring of the American market, it is foundational.
But the fact that a recording exists in Luminate does not automatically tell us whether it was AI-assisted.
A second layer of data is required.
Soundcharts
Soundcharts aggregates chart data from Spotify, Apple Music, Amazon, Deezer, iTunes, Shazam, TikTok, YouTube and other platforms, allowing users to track positions by country and over time.
It is excellent for answering:
“Where is this song performing, and how strongly?”
But not:
“How exactly was this recording created?”
Chartmetric
Chartmetric solves a similar problem by linking streaming, playlists, TikTok, Shazam, radio and other indicators around specific artists and tracks.
From the perspective of AI research, it is a valuable performance layer, but the provenance layer still needs to be added separately.
SIQA
A particularly interesting new independent project is SIQA, which maintains its own Top 100 AI Songs.
For the week beginning August 18, 2026, its highest positions included “RUBBERZ,” Xania Monet and other AI-oriented projects.
SIQA states that its methodology combines roughly equal components of Streams & Airplay and Social Impact, drawing from signals across Spotify, Apple Music, YouTube, TikTok, Instagram Reels, Shazam and other sources.
It also distinguishes between AI-Assisted, Human + AI Hybrid and Fully AI-Generated works.
This is a highly interesting experiment.
But SIQA should not be treated as equivalent to Billboard or IFPI.
It is an independent AI-specific ranking, not an official global industry chart.
Its greatest value is therefore as an additional research layer.
1212. Deezer Currently Holds One of the Most Valuable AI-Music Data Sets+
Deezer has gone further than most major streaming platforms.
The company developed proprietary detection technology designed to identify patterns associated with fully generated music, initially including content created using systems such as Suno and Udio.
During 2025–2026, Deezer began using AI detection to identify AI-generated recordings, label them, exclude them from editorial and algorithmic recommendations, and license the technology to third parties.
In June 2026, Deezer also introduced a tool for checking AI music in playlists created on other streaming services.
This marks an important transition.
Previously, a researcher often learned that a recording was AI-generated from an interview with the creator or from investigative journalism.
Now we are beginning to see the possibility of machine-level identification across entire catalogues.
But a detector should not become an absolute judge.
Every classification system has false positives, false negatives and difficulties caused by later audio processing.
The future is therefore likely to involve a combination of:
metadata + self-disclosure + distributor information + detection + human verification.
1313. The Most Important Change of July 2026: AI Enters Official Chart Infrastructure+
On July 30, 2026, IFPI announced international principles governing the eligibility of recordings developed with generative AI for official music charts.
In broad terms, an AI-developed recording may be eligible if:
- the AI service used is lawful and authorized;
- the work is substantially human made;
- there is no evidence of streaming or chart manipulation.
Additional considerations include copyright, personality rights, the terms governing the AI service, and appropriate disclosure of AI use.
IFPI has begun applying these principles across its own chart programs in Latin America, MENA, Africa and Southeast Asia, while working with more than twenty additional national chart programs.
This may be more important than any individual AI hit.
Because for the first time, the industry is formally stating:
AI as a tool does not automatically disqualify music from chart participation.
But also:
a fully synthetic recording and a recording in which AI assists a human creator should no longer automatically be treated as the same thing.
This distinction is likely to become one of the central boundaries around which the next model of the music industry is built.
1414. What Is Happening at Spotify?+
Spotify is also moving toward more detailed transparency.
The platform is developing AI Credits, through which the use of AI can be identified in areas such as lyrics, vocals, instrumental performance or production.
The relevant information is delivered through metadata supplied by labels and distributors.
In addition, on August 11, 2026, Spotify announced a separate AI Persona label, with rollout planned from mid-September 2026.
And here an important distinction must be made.
AI Persona answers the question:
Is the public artist identity itself an AI persona?
AI Credits answer a different question:
Where was AI used inside the production of this particular recording?
These are not the same thing.
A human artist can release an AI-assisted track.
An AI persona can have a human lyricist.
This is why future music metadata needs to describe individual creative roles inside individual works, rather than simply classifying an artist as a whole.
1515. Where Is the Real Competition Between Human and AI Music?+
Several conclusions can be drawn from the available evidence.
AI Is Not Yet Dominating Mass Consumption
Despite extraordinary production volume, Deezer currently sees only about 1–3% of streams going to fully AI-generated catalogue.
AI can generate enormous supply, but human attention does not scale at the same speed.
This distinction is fundamental.
But AI Can Already Produce Genuine Hits
Germany, the United States, Sweden and the United Kingdom all demonstrate that audiences are perfectly capable of listening to AI-generated music at substantial scale.
The technological assumption that “people will immediately recognize it and reject it” has effectively collapsed.
AI’s First Chart Victories Often Appear in Narrower Rankings
Digital Sales, Downloads, Viral and genre charts provide easier entry points.
Any serious research must therefore distinguish between:
No. 1 on iTunes,
No. 1 on Spotify Viral,
No. 1 on Country Digital Song Sales,
No. 1 on Spotify Top 50,
No. 1 on an official national singles chart,
No. 1 on the Billboard Hot 100.
These are six entirely different claims.
The Boundary Is Gradually Moving Toward the Major Charts
“RUBBERZ” reaching No. 58 on the Hot 100 is a much stronger market signal than reaching No. 1 on a small download ranking.
If such cases begin to repeat, we will no longer be talking about a separate AI niche.
We will be talking about penetration into mainstream consumption.
1616. The Most Interesting Shift: The Core Problem Is No Longer Generation Quality+
Only a few years ago, the central question was:
“Can AI make music good enough for people to listen to?”
Today, that question has largely been answered.
Yes.
The new questions are different:
Who is the author?
Which model was used?
What material was that model trained on?
What exactly did the human creator contribute?
Was AI use disclosed?
Was the model lawfully used?
Are the streams organic?
Should such a work compete in the same chart as a human-led recording?
The debate has therefore moved from generation quality to provenance, transparency, rights and measurement.
That is a profound shift.
1717. What This Could Mean for the Alex Kryve Music & AI Initiative+
This is where the opportunity emerges not merely for another article about artificial intelligence, but for genuine ongoing research.
Today we have Billboard.
We have IFPI.
We have Spotify Charts.
We have Soundcharts and Chartmetric.
We have Deezer AI Detection.
We have emerging projects such as SIQA.
But what is still largely missing is a single independent layer that connects:
popularity + chart type + degree of AI use + evidence of provenance + human contribution.
That creates a clear opportunity for the Alex Kryve Music & AI Initiative to establish its own AI Music Chart Observatory.
Not another “Top 100 Best AI Songs.”
A research database.
For every identified AI-involved release, the Observatory could record:
ISRC; artist and song title; country; specific chart; current position and peak position; chart category — streaming, sales, radio, combined or viral; AI-Generated / AI-Assisted / disputed; source of AI-provenance evidence; tools used; known human contributors; disclosure status; official chart eligibility; and any confirmed concerns or rulings involving stream manipulation.
After one year, this would no longer be merely an article.
It would be a dataset.
And from that dataset, several original indicators could be created.
Mainstream AI Penetration Index — how many AI-involved recordings actually enter major national all-genre charts.
AI-Native Performance Index — which AI-oriented projects generate real consumption and social traction.
Human + AI Transparency Index — how consistently the market discloses the use of generative technology.
Such a resource could become valuable not only to independent artists, but also to researchers, journalists, distributors, labels, rights organizations and AI platforms themselves.
1818. What Such a Database Could Tell Us One Year from Now+
Today, we can point to individual examples.
After a year of systematic monitoring, we could begin answering much more serious quantitative questions.
For example:
Is the number of AI-assisted tracks entering Billboard growing faster than the number of fully generated tracks?
Which genres provide the easiest path for AI into mass audiences?
Does AI enter streaming charts, or remain concentrated in downloads?
Which territories adopt AI music fastest?
What happens to AI-related tracks after mandatory disclosure is introduced?
Does an AI label influence listener behavior?
What proportion of AI-generated recordings exists among uploads compared with their share of actual streams?
And perhaps most importantly:
does AI improve or reduce the ability of independent human creators to compete with the traditional music industry?
That last question deserves particular attention.
Because AI should not be understood only as a machine producing competing content.
It may also function as a production multiplier for human creators.
From the perspective of the independent artist economy, that is an entirely different model.
1919. Main Conclusion+
As of August 2026, it would be incorrect to say:
“AI music has taken over the charts.”
The available data does not support that conclusion.
But it would be equally incorrect to say:
“AI music exists somewhere outside the real music market.”
That is no longer true either.
AI-related music has already reached No. 1 on platform charts.
It has already reached No. 1 on individual Billboard rankings.
It has already appeared in official national charts.
It has already reached the Billboard Hot 100.
And at the same time, official industry organizations have begun creating rules defining which forms of AI-involved music should be allowed to participate in official charts at all.
But perhaps the most striking number today is still this:
>50% of new uploads — 1–3% of streams.
At least on Deezer.
It tells us that the main battle of the coming years will not be about the ability to create music.
That ability is becoming almost unlimited.
The battle will be for a far scarcer resource:
human attention.
And in a world where machines can produce one hundred thousand songs, perhaps human story, intention, identity and meaning will become not less valuable, but more valuable.
AI is changing the way music is produced.
But the market does not ultimately measure the number of files created.
It measures what people choose to listen to again.
Key Sources and Databases for Further Monitoring
IFPI — new AI-Generated / AI-Assisted framework
IFPI — AI eligibility principles for official charts, July 30, 2026
Deezer — AI exceeds 50% of daily music uploads
Official Charts — Eddie Dalton / Another Day Old
Associated Press — Breaking Rust / Walk My Walk
Sveriges Radio — AI hit that reached No. 1 on Spotify but was excluded from Sweden’s official chart
The Verge — investigation into the origin of “RUBBERZ”
For ongoing monitoring, the most useful sources also include Billboard/Luminate, IFPI, Spotify Charts, Apple Music/iTunes charts, Official Charts Company, Soundcharts, Chartmetric, Deezer and SIQA.
No single one of them currently provides the entire picture.
It is precisely by combining them that we can begin to distinguish between the enormous flow of AI-generated content and something much more important:
a real AI music market.