Treblo’s AI music detector intensifies the Fenix Flexin debate over ‘Rubberz’

Fenix Flexin’s song “Rubberz” is now at the center of a sharper AI controversy after Treblo released an open-source classifier that says the track is “very likely Treblo” with high confidence. The new tool does not prove the song was generated by AI, but it gives the strongest public signal yet that Treblo’s model may have helped create the release.

The development matters because the track has already sparked one of the most visible mainstream disputes over AI in music, raising questions about authenticity, disclosure, and how listeners can tell whether a hit song was human-made, machine-made, or a blend of both.

What began as online suspicion has turned into a technical and reputational problem for the artist, with Treblo effectively backing the idea that its system was involved while Fenix Flexin continues to deny using AI.

What happened with Treblo and Fenix Flexin?

Treblo published a new open-source AI music classifier on Monday that is designed to detect whether a track was created with Treblo’s own generation model. The company says the tool is not built to identify music made with other AI systems, only audio that appears to have come from Treblo.

According to Treblo, the classifier has a very low false-positive rate, which the company describes as less than one in 1,000. That is not the same as perfect certainty, but it is enough for Treblo’s system to flag “Rubberz” as a likely match.

The company said its classifier marked the song as “very likely Treblo” and did so with high confidence. In practical terms, that means Treblo itself is now publicly leaning into the possibility that its model played a major role in producing the track.

Why the result matters

The significance goes beyond one song. If the classifier is accurate, “Rubberz” could become one of the clearest examples of an AI-generated song making a serious commercial run, including the possibility of reaching the Billboard Hot 100.

That possibility is what makes this story so combustible. The music industry has spent the past two years wrestling with generative AI in nearly every corner of its workflow, from vocal cloning and stem separation to full-track generation. A mainstream single attached to a recognizable artist forces that debate out of the abstract.

It also creates a new challenge for labels, streaming platforms, and listeners: what happens when a song sounds plausible, performs well, and still leaves no obvious trail of how it was made?

How did the controversy around “Rubberz” begin?

The debate started with skepticism from listeners and observers who thought “Rubberz” had the telltale signs of AI production. The suspicion did not stay confined to internet chatter for long.

Musician Medasin, who has followed the discussion closely, was especially confident that the song had been made using Treblo rather than a different tool. That view now appears to have been reinforced by the company’s own classifier.

Before Treblo’s announcement, the song’s production was already being dissected online, with critics arguing that the audio did not sound like a straightforward human recording session. The new classifier has not settled every factual question, but it has given those claims more technical weight.

Treblo’s chief executive, Ryan Tremblay, said the classifier returned a very high probability score for “Rubberz” and suggested the released audio was generated almost entirely by the company’s model.

Tremblay framed the result as potentially historic, saying that if the track was made that way, it would be the first known AI-generated song to reach the Billboard Hot 100. He also said that someone appears to have used Treblo’s model to create a sound that connected with a large audience.

What is Treblo’s AI music classifier?

The classifier is Treblo’s attempt to identify whether a piece of music was made using its own generation system. Unlike a general AI detector, it is not intended to judge every possible AI production method; it is specifically trained around Treblo output.

That narrow focus makes the tool more useful as a platform-specific signal, but it also means the system should not be mistaken for a universal AI lie detector. A track that comes back clean would not necessarily mean it was made entirely by humans, and a flagged track does not automatically prove unlawful or undisclosed use.

Still, for a case like “Rubberz,” the classifier provides a new layer of evidence. If Treblo’s own system says a song is highly likely to have originated from Treblo, that is hard for the company to walk back and hard for the artist to ignore.

How reliable is the detection?

Treblo says the classifier’s false-positive rate is below 1 in 1,000, which suggests a high level of precision on paper. But any detector can be fooled by unusual audio, incomplete files, post-processing, or adversarial attempts to disguise generation traces.

That means the tool should be read as strong supporting evidence, not a final forensic judgment. In a dispute involving public denial, however, even a strong probability score can shift the burden of explanation.

In this case, the company’s own output may be more damaging than any outside critique, because the source of the suspicion is now the system that allegedly produced the song in the first place.

Why is Fenix Flexin denying AI use?

Fenix Flexin has continued to reject claims that “Rubberz” was created with AI. The artist has also posted short clips that he says show project files from the recording session, apparently to prove that the song was assembled through conventional studio work.

Those clips have not ended the dispute. Critics say the videos are too low in quality to settle anything, and they argue that the materials do not clearly explain how the finished track was built.

In a particularly pointed accusation reported by Stereogum, Medasin suggested in an Instagram story that Fenix may have used an AI stem separator to fake a session. That story later expired, but the allegation added another layer to the public debate.

As of the latest reports, Fenix had not responded immediately to a request for comment. The silence is notable because the conversation is not just about one song’s production credits; it is also about whether an artist should be expected to disclose the use of generative tools when listeners hear something as a finished record.

What are the broader stakes for artists?

The stakes are now bigger than one controversy. If artists can release AI-assisted songs without clear disclosure, the line between performance and fabrication becomes harder to police.

That matters for several reasons:

• fans may feel misled if they believe they are supporting human performance;
• labels may face pressure to verify credit and provenance;
• streaming services may need tools to label or detect AI content;
• songwriters and producers could face new questions about creative authorship;
• chart success may become a flashpoint if AI songs compete directly with human-made releases.

“Rubberz” has become a test case because it sits at the intersection of virality, suspicion, and technical evidence. If the track becomes a hit while the AI questions remain unresolved, the music business may be forced to confront a problem it has mostly discussed in theory until now.

How does this fit into the wider AI music boom?

This story arrives as generative music tools are becoming more capable and more visible. AI systems can now mimic voices, assemble instrumental tracks, and imitate genre conventions with surprising speed. For creators, that means lower barriers to production. For rights holders and listeners, it raises difficult questions about ownership, originality, and consent.

Several tensions are now colliding at once:

1. AI music creation is getting easier and more convincing.
2. Detection remains imperfect and often platform-specific.
3. Public trust is fragile when artists deny using synthetic tools.
4. Commercial success can make a questionable release harder to challenge.

That combination explains why Treblo’s classifier release immediately drew attention. It is not just a product announcement. It is also a strategic move in a larger fight over how AI music will be identified, marketed, and judged.

Why the Billboard angle matters

The Billboard angle matters because chart placement gives the controversy cultural weight. A song can be dismissed as an online curiosity until it starts competing in the same commercial arena as mainstream pop, hip-hop, and rap.

If “Rubberz” were to become the first clearly AI-generated song to enter the Billboard Hot 100, the industry would have a benchmark moment comparable to other technological shocks in music history. Even if that milestone never happens, the mere possibility helps explain why the story has drawn such intense scrutiny.

What does Treblo gain from this moment?

Treblo appears to gain visibility, validation, and leverage. The company’s detection tool not only spotlights its technical confidence, it also implicitly showcases the power of its generation model if the classifier is correct.

Ryan Tremblay’s comments suggest the company sees the episode as evidence that its technology can produce music listeners actually want to hear. That is a notable stance: rather than treating the controversy as a liability, Treblo is presenting it as proof that AI music can resonate in the real world.

Treblo’s CEO argued that the reaction shows someone used the company’s model to create a sound that connected with millions of people, and he cast that reaction as a positive sign for the technology.

That framing may appeal to investors, technologists, and some creators, but it will likely frustrate artists and critics who think the company is effectively celebrating an undisclosed AI hit.

How the evidence is stacking up

No single fact has ended the debate, but several elements now point in the same direction. The original suspicion, Medasin’s public confidence, Treblo’s own classifier result, and the weakness of the materials Fenix has posted all make the denial harder to sustain.

At the same time, the report does not provide a courtroom-style proof. There is no public session file audit, no independent forensic review, and no formal ruling from a platform or rights body. Instead, the story is evolving through a mixture of technical signals and public argument.

That is increasingly how AI disputes are being decided in the open: not through a single authoritative answer, but through a growing pile of probabilistic evidence that shapes public belief long before any official process catches up.

What happens next?

The next phase will likely depend on whether additional evidence emerges, whether Fenix offers a more detailed explanation, and whether Treblo continues to stand behind the classifier result in public.

If the song moves further up the charts, the scrutiny will only grow. If more artists begin experimenting with Treblo or similar tools, the “Rubberz” episode may become a reference point for how detection, denial, and disclosure work in the AI era.

For now, the central issue is not simply whether one song was made with a generator. It is whether music audiences are entering a period where the source of a hit can be algorithmic, contested, and commercially successful all at once.

That question is no longer theoretical. With Treblo’s own classifier pointing toward its model, the debate around “Rubberz” has shifted from rumor to a live case study in the future of AI music.

Background: why AI music disputes keep escalating

The “Rubberz” controversy fits a broader pattern that has emerged across entertainment and technology. As generative tools improve, the gap between what audiences hear and what creators admit becomes more difficult to manage.

There are three reasons these disputes keep escalating. First, music can be difficult to authenticate after the fact because many production steps leave no obvious trace in the final file. Second, AI tools are increasingly modular, meaning a song may be partly generated, partly edited, and partly human-performed. Third, there is little industry consensus on how to label or disclose AI-assisted work.

That uncertainty creates space for misunderstandings, strategic denials, and genuine confusion. A song can be accused of being AI because it sounds synthetic, even when the evidence is thin. But when a company’s own detector points toward its model, the argument changes from vibe-based speculation to a technical question with major reputational consequences.

For artists, this may force a new standard of transparency. For companies, it may demand stronger provenance tools. And for listeners, it could mean learning to treat the sound of a track as only part of the story behind it.

In the end, what does this case tell us?

This case shows that the fight over AI in music is no longer about hypothetical demos or experimental clips. It is about released songs, public denials, technical classifiers, and the possibility that an AI-generated track could break into the mainstream under a human artist’s name.

Whether “Rubberz” was fully made with Treblo, partly assisted by AI, or assembled in some more complicated way, the controversy has already achieved one thing: it has forced the industry to confront how little certainty exists once synthetic tools enter the studio.

And if Treblo is right, the bigger story is not only that AI made the song. It is that AI may already be good enough to make a commercially viable hit while everyone argues about who, or what, actually performed it.