Fenix Flexin finally admits using Treblo AI to make ‘Rubberz’ after producer exposed the tool

An LA rapper spent months dodging questions about his 80s synth-pop hit, then finally cracked when a producer showed the receipts.

Fenix Flexin’s admission that he used Treblo—an AI music generation tool formerly called Sonauto—to create “Rubberz” marks a turning point in how mainstream artists are reckoning with machine-generated sound. The song charted on the Billboard Hot 100. Millions of listeners streamed it without knowing its origin. Now, as AI tools make it trivially easy to produce radio-ready tracks, the question isn’t whether artists will use them. It’s whether they’ll ever admit it.

The unraveling started when producer Medasin published videos claiming Treblo generated “Rubberz.” Treblo’s creators then released an AI detector—essentially a fingerprint scanner for their own tool—that flagged the song as machine-made. The evidence was airtight. Fenix’s denials collapsed.

On Instagram, responding to a comment that compared his situation to rapper Tyga’s admission about using AI on his 80s-influenced album $tarface, Fenix wrote: “never said I didn’t use AI, i said it had nun to do w recording process. I didn’t even know what da app was till we mixed.” The framing is careful. He’s not claiming innocence. He’s redefining what “using AI” means—separating the tool from the creative act, as if Treblo was just another plugin, like reverb or compression. For a broader look at how AI-generated music is already reshaping critical evaluation, the pattern of denial and reframing is not unique to Fenix.

What Does Treblo Actually Do—and Why Does It Matter?

That distinction matters because it reveals the fault line running through the music industry right now. Treblo doesn’t just tweak existing recordings. It generates the entire composition from scratch—melody, harmony, drums, bass, synth lines. It’s not a filter. It’s a songwriter. Yet Fenix’s comment suggests he’s treating it as something closer to post-production, a technical step that happens after the “real” creative work is done.

The tool itself is telling. Treblo markets itself as a music generation platform that can create full tracks in specific genres and styles. Users input parameters—tempo, mood, instrumentation—and the system outputs a finished song. No human musician required. The cover art for “Rubberz,” according to The Verge’s reporting, is “almost certainly AI” too, suggesting a fully automated pipeline from concept to streaming release.

Why Did the Denial Last So Long?

What makes this moment significant isn’t that an artist used AI. It’s that an artist used AI to reach the Billboard Hot 100, then lied about it, then only admitted it when caught. That sequence matters because it shows the incentive structure at work. There’s no penalty for using AI if no one finds out. Streaming platforms don’t flag AI-generated music. Listeners don’t know. Charts don’t care. The only cost is if someone exposes you—and even then, as Fenix’s Instagram comment shows, you can reframe the narrative and move on.

Tyga’s earlier admission about $tarface set a precedent. He leaned into it, treating AI as a legitimate creative tool rather than something to hide. But Tyga’s transparency came after the fact, not before. He didn’t volunteer the information. He responded to criticism. Fenix followed the same arc: deny, minimize, reframe. The pattern suggests that artists see AI disclosure as a liability to be managed, not a fact to be celebrated upfront.

Is the Music Industry’s Transparency Problem Structural?

Treblo’s AI detector is its own strange artifact. The company built a tool to identify its own product’s fingerprints—essentially creating a weapon that could be used against artists who use it. That’s either a calculated move to build trust (“We can prove authenticity”) or a sign that the company knows the tool’s output will eventually need verification. Either way, it’s an admission that AI-generated music will become indistinguishable from human-made music, and we’ll need machines to tell the difference.

For listeners, the implications are immediate. You’ve probably heard AI-generated music on streaming platforms without realizing it. You might have paid for it, added it to playlists, or shared it with friends. “Rubberz” reached millions of people. How many of them would have engaged with it differently if they’d known its origin? We don’t know because artists aren’t required to disclose. The industry has no standard for transparency. This absence of accountability has a structural parallel worth noting: the same logic that allowed behavioral data to be harvested and deployed without user awareness during the Cambridge Analytica era—where the absence of disclosure requirements was treated as permission—is now operating in the music economy. When platforms profit from engagement regardless of origin, and creators face no disclosure mandate, the default becomes concealment.

The 80s synth-pop aesthetic that both “Rubberz” and Tyga’s $tarface adopt is itself revealing. Nostalgic, retro-futuristic, slightly kitschy—it’s a style that feels safe for AI to attempt. The genre has clear sonic rules, established production techniques, and a limited palette of sounds. Treblo can learn those patterns and replicate them convincingly. A systematic review of generative AI published in ScienceDirect confirms that music generation systems trained on deep learning architectures perform most reliably within genres defined by consistent structural rules—precisely the conditions that make retro-pop an ideal testing ground. Whether the tool could generate something genuinely novel, something that breaks genre conventions or surprises listeners, is still an open question.

What Happens When Platforms Profit From Ambiguity?

What happens next depends on whether the music industry treats this as a scandal or a signal. If Fenix’s admission becomes a cautionary tale—a warning about getting caught—then artists will simply get better at hiding AI use. If it becomes a conversation about what AI-generated music means for authenticity and fair compensation, then we might see actual policy changes. The broader question of platform liability in digital content is already being contested in courts, and music may be the next domain where disclosure obligations become a legal rather than voluntary matter.

The streaming economy rewards volume and engagement, not origin. An AI-generated song that reaches the Hot 100 generates the same royalties and streams as a human-made one—though the question of who actually receives those royalties when AI is involved remains murky. Record labels have financial incentive to use Treblo and similar tools. Artists have career incentive to hide it. Listeners have no mechanism to know. That misalignment is where the real story lives.

Fenix Flexin’s Instagram comment—”I didn’t even know what da app was till we mixed”—is either honest or a masterclass in plausible deniability. Either way, it works. The admission defuses the scandal without requiring him to apologize or change anything. He used Treblo. He’s not denying it anymore. Life moves on. The next artist will face the same choice: disclose or deny. And based on the incentives in place, most will probably choose to deny until they’re forced to admit.