AI Music Detector Puts Fenix Flexin’s ‘Rubberz’ Under a New Spotlight

An AI music app has publicly flagged Fenix Flexin’s viral summer track “Rubberz” as likely machine-generated, intensifying questions that have followed the song since its June release. The same detector also marked a Tyga track as likely made with the app, raising the stakes in a growing debate over how listeners can tell whether a hit song was built by a human, a model, or both.

The latest twist matters because it could turn a messy fan dispute into a broader industry moment: AI tools are no longer just making music, they are also starting to police music. That puts labels, artists, and listeners in the middle of an authenticity fight that is quickly becoming one of the defining arguments in modern pop and rap.

What began as speculation around a strange-sounding rap single has now evolved into a story about detection, disclosure, and the limits of digital proof. And while no one has produced definitive public evidence that “Rubberz” was generated by AI, the conversation has shifted from rumor to a more formal test of competing claims.

Fenix Flexin and producer Purps on the Beat have denied using AI, even as producers and online observers have pointed to unusual vocals, stylistic oddities, and apparent metadata clues. The app at the center of the latest controversy, Treblo, now says its own detector thinks the song was made with its model, giving critics of the track a new talking point — and supporters of AI music a new sign that the technology is being normalized faster than many expected.

What happened with “Rubberz”?

“Rubberz” became a summer controversy because it sounded unlike a typical Fenix Flexin release and because it appeared to fit a pattern of AI-assisted music creation that listeners increasingly recognize.

The song, released in June, climbed into the Billboard Hot 100 and drew millions of views for its accompanying video. It also prompted immediate confusion: despite Fenix’s reputation in rap, the record leaned hard into a bright, synth-heavy, ’80s-inspired style that sounded more like a retro pop experiment than a straightforward West Coast rap single.

Listeners also noticed something else: the voice on the song did not appear to match the artist they knew. Fenix, who rose to prominence with Shoreline Mafia, had been associated with a recognizably nasal and rougher delivery. On “Rubberz,” the vocal tone seemed cleaner, lighter and more polished, which fueled the idea that the track might not be entirely human-made.

Both Fenix and Purps repeatedly denied the allegations. Fenix said he wrote the lyrics himself and suggested that the track was simply recorded with heavy processing. Still, the skepticism never fully went away, and the debate escalated as the song kept climbing.

Why did the controversy spread so quickly?

The controversy spread because the song sat at the intersection of three hot-button issues: AI music, artist authenticity and viral fandom. When a track is catchy enough to travel widely but strange enough to trigger suspicion, the internet tends to turn it into a referendum on the future of music itself.

Part of the intrigue was that no major AI detector initially confirmed the accusations. That meant the public was left with circumstantial evidence: a voice that sounded altered, a genre pivot that seemed abrupt, and lyrics that some listeners thought resembled patterns generated by synthetic music tools.

Meanwhile, the song kept gaining traction. That created a paradox that has become common in the AI era: a piece of content can be controversial, unverified and massively successful all at the same time. The more people argued about “Rubberz,” the more visible it became.

Music producer Medasin, who later posted a breakdown of the track online, argued that the resemblance between the disputed song and Treblo-generated samples was too strong to ignore, describing the overlap as more than coincidence.

That claim is the one that pushed the debate into a new phase, because it did not merely say the song sounded suspicious. It attempted to identify the specific tool that may have been used to create it.

How did Treblo become part of the story?

Treblo entered the conversation after a producer said he had recreated a song strikingly similar to “Rubberz” using the app. Treblo was not one of the biggest music-generation products in the market, which made the allegation more surprising and arguably more believable to those who had dismissed earlier speculation.

According to the accusation, Treblo — previously known as Sonauto — had changed names shortly before “Rubberz” arrived. The producer who made the claim pointed to a screenshot from Fenix’s social media that appeared to show a file name tied to the earlier Sonauto branding, suggesting a possible link between the artist’s teaser material and the tool.

But the strongest part of the argument was not the name change or a screenshot. It was the demonstration. By using prompts associated with the “80s synthwave” style and asking for rap lyrics, the producer said he was able to make a song with similar phrasing, structure and sonic texture.

The comparison caught attention because of specific lyric overlaps and recurring motifs. According to the breakdown, the generated songs and “Rubberz” both referenced jewelry, money, pockets and glass imagery, along with a repeated pattern of rhyming and thematic movement from nightlife to daybreak. Those parallels do not prove the same song was copied, but they do show why listeners were uneasy.

What did the Treblo detector do?

The Treblo detector publicly flagged “Rubberz” as “Very likely Treblo,” giving the dispute a technical answer, even if not a definitive one.

After the controversy intensified, Treblo launched a page asking whether a song was made with its system. The company said the detector was designed to identify tracks generated with its model with little or no modification, and it encouraged listeners to consider disclosure and transparency.

When “Rubberz” was tested, the result came back as highly likely to have been made with Treblo. A Tyga track was also flagged as likely Treblo-generated. The company says its false-positive rate is under one in 10,000, although that claim has not been independently verified.

The detector is notable not just because it exists, but because Treblo released the code under an open-source license. That means outside researchers can inspect how it works, test it and challenge its performance. In a field where many AI tools are opaque, that level of openness may become a model for others — or a target for criticism if the detector proves unreliable.

Who is Medasin, and why does his video matter?

Medasin is an independent producer whose video turned private speculation into a public accusation. In stories like this, the messenger can matter nearly as much as the evidence.

Rather than simply posting a hot take, Medasin said he tested the app himself and compared the results to Fenix’s song. He later told WIRED that he believed the match was essentially impossible to ignore. His argument was that the same phrasing patterns and melodic instincts kept appearing when he used Treblo in a certain way, suggesting that “Rubberz” may have come from the same pipeline.

He also said he had sat on the findings for about a month, expecting someone else in the music press to connect the dots. That never fully happened, which helps explain why his video resonated: it packaged a messy, technical, internet-native rumor into a shareable explanation.

The reach of the video also reflected a familiar dynamic in online music culture. Fans often tolerate gossip, but they pay closer attention when a producer shows the work. Whether or not the conclusion is correct, a methodical breakdown carries more weight than a vague accusation.

What happened after Fenix posted project-file evidence?

Fenix tried to counter the accusation by posting what he said was the original project file for “Rubberz,” but the move quickly backfired. Instead of calming the debate, it drew fresh scrutiny from producers who suspected the file might not be what it seemed.

Commenters and music makers began arguing that the screenshot or screen recording resembled a stem-separated file rather than an untouched studio session. A stem separator is a tool that breaks a finished song into individual elements such as vocals, bass, drums and instrumental layers. If that theory were true, it would undercut the idea that the file was proof of a fully human-made original session.

Shortly after the backlash, the post was removed. Neither Fenix nor Purps has publicly explained why it disappeared or responded in detail to the renewed allegations. That silence has only deepened the impression that the issue remains unresolved.

In online music disputes, deleted posts often become part of the story themselves. Here, the removal of the file did not stop the rumor; it became fuel for it.

Why does this matter beyond one song?

This matters because it shows the music industry is moving from guessing whether something is AI-made to building tools that can say so. That shift could change how songs are marketed, monetized and trusted.

Streaming services, social platforms and publishing tools are already experimenting with flags and filters for synthetic content. The broader trend is clear: platforms are being pushed to tell users when text, audio, images or video may have been created by a machine. In music, that pressure is especially intense because the line between production technique and authorship has always been blurry.

For some listeners, AI use is not automatically a problem. They care about the final result, not the process. For others, disclosure is essential because a song’s emotional value depends on whether the voice and writing come from lived experience. That tension is now moving from niche debate to mainstream product design.

There are also commercial consequences. A song flagged as synthetic may raise questions about royalties, brand deals, chart eligibility and platform monetization. If a major hit can be generated, marketed and disputed without clear disclosure, labels may face stronger pressure to document how tracks are made.

How are platforms responding to AI content?

Platforms are responding by adding labels, detectors and reporting tools, although their approaches vary widely. Some services are trying to warn users; others are trying to filter or limit synthetic content.

• LinkedIn introduced a feature allowing users to mark posts that appear to be AI-generated.
• Substack rolled out a tool meant to detect AI-written newsletters.
• Tidal has taken steps to block monetization of music created with AI.
• Spotify does not have a formal AI tag, but users can report suspicious tracks through crowdsourced trackers.

These moves reflect a wider anxiety about authenticity in digital media. The concern is not only that AI can create realistic content, but that it can do so at scale, making it harder for audiences to know what is real and what is synthetic.

What role could regulation play?

Regulation could force more consistent disclosure, but it is unlikely to settle the artistic debate on its own. Lawmakers are beginning to frame AI labeling as a consumer-protection issue.

In the United States, senators from both parties have floated the idea of an AI Labeling Act that would require platforms to visibly mark machine-generated content. Supporters argue that people have a right to know whether what they are consuming came from a human or an algorithm. Critics worry that rigid labeling rules could be hard to enforce and easy to game.

Even if legislation passes, it would not answer the harder cultural questions: What counts as AI-assisted versus AI-generated? How much editing makes a song “human” again? And does the audience care once a track becomes a hit?

Those questions are why the “Rubberz” controversy matters. It is not just about one single song or one rapper’s reputation. It is about where the industry draws the line between tool use and creative authorship in a world where generative software can imitate style with growing precision.

How did fans react?

Fans reacted with a mix of suspicion, indifference and enthusiasm. That split is one reason AI music has become such a difficult issue for the industry to address.

Some listeners immediately dismissed the song once AI accusations appeared. Others said they liked it regardless and saw no reason to object if the music sounded good. A third group treated the dispute like a fun internet puzzle, focusing less on ethics than on whether the detective work was clever.

This range of reactions points to a basic truth: audience expectations are changing faster than the institutions around them. Many listeners already consume heavily processed vocals, ghostwritten lyrics, sampled compositions and digitally altered performances without thinking too much about authorship. AI simply adds a new layer.

That may be why Medasin’s argument struck such a nerve. He framed the issue not as a purity test, but as a transparency test. In his view, artists should be free to use whatever they want, but they should not mislead fans about how a song was made.

Medasin said the broader issue was not whether AI belongs in music, but whether audiences deserve honesty about its use so they can judge the work for themselves.

Timeline of the “Rubberz” dispute

Here is a simplified timeline of the main developments in the controversy.

What does Treblo gain by flagging its own users?

Treblo gains credibility by turning its detector into a transparency product, even though that move is commercially awkward. A company that can identify its own output may be able to position itself as responsible, not just powerful.

At the same time, flagging songs made with its model could raise questions about customer privacy, artist confidentiality and product trust. If creators use a music app believing the output is anonymous or difficult to trace, a built-in detector changes that expectation.

Still, from a PR perspective, Treblo’s response is savvy. Rather than hiding from the controversy, it has effectively said: if our tool is being used to make hits, we can also help identify them. That stance may appeal to platforms and regulators seeking some sort of verification layer.

Whether musicians will see that as helpful or hostile is another matter. For artists experimenting with generative tools, being detectable may feel like a warning sign. For labels worried about compliance, it may feel like a relief.

How reliable is the evidence?

The evidence is suggestive, but it is not conclusive in the public record. That is the most important caution in the entire story.

There are several pieces of circumstantial support for the AI theory: the sudden style shift, the unusual vocal tone, the apparent file naming clue, the lyric similarities, the producer demonstration, and the Treblo detector’s own result. But none of these alone proves the final answer beyond doubt.

Independent review is especially important in stories like this because AI detectors can produce false positives, prompts can converge on similar phrasing by coincidence, and artists can emulate styles deliberately without machine assistance. The public still does not have a definitive forensic breakdown of the track’s full production history.

What the available material does show is that the music industry is entering a phase where suspicion can be algorithmically reinforced. The detector does not just resolve the dispute; it becomes part of the dispute.

Why the argument may never fully end

The argument may never fully end because AI music disputes mix technical ambiguity with cultural preference. People are not just debating evidence; they are debating what counts as legitimate art.

Some fans will accept AI involvement if the song is entertaining. Others will view undisclosed machine assistance as a breach of trust. Both positions can coexist, which means no single detector or denial will settle every listener’s view.

That is the uncomfortable reality of synthetic media. The tools are improving, the detection tools are improving, and the public is being asked to decide how much disclosure matters. The “Rubberz” episode shows that the answer is not obvious.

For now, the song remains a hit, the accusations remain disputed, and Treblo’s detector has added a new layer to an already messy story. What began as a question about one summer track is now part of a much larger conversation about how the music business will handle AI in the years ahead.

And that may be the real headline: not whether one song was made by a machine, but that the machinery for proving it is now arriving almost as quickly as the songs themselves.

Key facts at a glance

As the debate continues, the core issue remains unchanged: if AI can help create music that listeners embrace, how much does the industry owe them in the way of disclosure? “Rubberz” has become one of the clearest tests yet of that question.