You're not really choosing if you don't know what you're listening to
Across the streaming landscape, machines now speak where humans once did — nearly four in ten new podcasts are generated entirely by artificial intelligence, a quiet saturation that mirrors upheavals already underway in music, film, and literature. The efficiency is undeniable, but so is the cost: when audiences can no longer tell the human voice from the synthetic one, something essential about the relationship between creator and listener begins to dissolve. This moment asks an old question in a new register — not merely who made this, but whether it matters that we know.
- Nearly 40% of new podcasts are now machine-made, with firms like Inception Point AI producing thousands weekly and claiming tens of millions of listeners — a scale no human creative team could match.
- The same flood is reshaping music, with over 50,000 AI-generated tracks hitting Deezer alone each day, and surveys confirming that most listeners can no longer tell the difference — and are unsettled by that fact.
- The deeper disruption is epistemic: audiences are losing the ability to make informed choices about what they consume, unable to know whether a voice belongs to a person or a process.
- Certification bodies like Proudly Human are emerging to verify human authorship, framing transparency not as a courtesy but as a fundamental right in an increasingly synthetic media environment.
- Pressure is building on streaming platforms to mandate AI labeling before the line between human and machine creation fades beyond recovery.
Somewhere in the catalog of a major streaming platform, a synthetic voice is discussing gardening tips or financial advice — created by a machine in minutes. According to the Podcast Index, nearly four in ten podcasts uploaded in recent months were generated entirely by artificial intelligence, a phenomenon the industry has taken to calling "podslop."
The mechanics are simple: a creator supplies a topic, a tool generates a script, assigns it to a synthetic voice, and uploads the result. No host, no studio, no editing required. Companies have built entire businesses on this model. Inception Point AI produces thousands of such podcasts weekly, marketing AI personalities — an English gardener, a financial advisor, a celebrity gossip commentator — each tailored to specific audiences. The firm claims over ten million listeners and calls its work "the next chapter in storytelling."
The pattern extends well beyond podcasting. Deezer alone receives over fifty thousand AI-composed songs daily. A survey of nine thousand people across eight countries found that most could not distinguish AI-generated music from human-made songs — and were troubled by their own inability to tell the difference.
The problem runs deeper than preference. As synthetic content multiplies, audiences lose the ability to make informed choices about what they consume. Certification bodies like Proudly Human have emerged to verify human authorship, with founder Trevor Woods framing the issue as a matter of fundamental rights. Without clear labeling, human creators lose the ability to signal their investment, and audiences lose the ability to support them.
What began as a tool for efficiency now risks becoming a form of noise that obscures rather than amplifies human creativity. The question is whether platforms will act to restore that clarity before the distinction becomes too faint to recover.
Somewhere in the vast catalog of streaming platforms, a synthetic voice is right now discussing gardening tips, financial advice, or celebrity gossip—and it was created by a machine in minutes, not hours or days. This is the new reality of podcasting. According to data from the Podcast Index, nearly four out of every ten podcasts uploaded to streaming services in recent months were generated entirely by artificial intelligence, a phenomenon the industry has begun calling "podslop."
The mechanics are straightforward enough. Free online tools now exist that can produce a complete podcast episode in minutes. A creator supplies a topic, the system generates a script, assigns it to a synthetic voice, and uploads the finished product. No human host required. No studio time. No editing. Companies have built entire business models around this efficiency. Inception Point AI, for instance, produces thousands of these podcasts each week. The company markets a roster of AI personalities—an English gardener named Nigel Thistledown, a financial advisor called Pennie Power, a celebrity gossip commentator known as VV Steele—each designed to appeal to specific audience segments. The firm claims to reach more than ten million listeners and describes its work as "the next chapter in storytelling."
This explosion of machine-made content is not confined to podcasting. The same pattern is unfolding across creative industries. Music streaming services are drowning in AI-generated tracks. Deezer alone receives over fifty thousand AI-composed songs daily, accounting for roughly a third of all new music uploaded to the platform. Film and literature face similar pressures. A survey conducted last year by Ipsos and Deezer, spanning nine thousand people across eight countries, found that the vast majority of respondents could not distinguish between AI-generated music and songs created by human musicians. More troubling than the inability itself was the discomfort it produced. People did not like not knowing the difference.
The authenticity problem cuts deeper than mere preference. As synthetic content proliferates, audiences lose the ability to make informed choices about what they consume. They cannot reliably know whether they are listening to a human creator or a machine. This asymmetry of knowledge has prompted calls for transparency. Some companies now offer certification services designed to verify that content was made by humans. Trevor Woods, founding CEO of Proudly Human, an international certification body, frames the issue as a matter of fundamental rights. "Providing the clarity to confidently choose human-created content is a fundamental right in an era where AI is increasingly masquerading as human," he said. The organization positions itself as the only international certification mark verifying human authorship across multiple creative industries.
The pressure on streaming platforms to implement clear labeling of AI-generated content is mounting. Without such measures, the distinction between human and machine creation risks becoming invisible to the average listener or viewer. Creators who invest time and skill in their work lose the ability to signal that investment to audiences. Audiences lose the ability to support the creators they value. The market becomes harder to navigate, not easier. What began as a tool for efficiency—the ability to generate content quickly and cheaply—threatens to become a form of noise that obscures rather than amplifies human creativity. The question now is whether platforms will act to restore that clarity before the distinction becomes too faint to recover.
Notable Quotes
Providing the clarity to confidently choose human-created content is a fundamental right in an era where AI is increasingly masquerading as human— Trevor Woods, founding CEO of Proudly Human
The Hearth Conversation Another angle on the story
Why does it matter if a podcast is made by AI or a person, if the content is useful?
Because you're making a choice every time you listen. If you don't know you're listening to a machine, you're not really choosing—you're being deceived. The person who made the podcast, if it was a person, deserves credit for that work. The listener deserves to know what they're supporting.
But these AI tools are free. Doesn't that democratize podcasting for people who couldn't afford to produce content before?
It does, in a narrow sense. But when thirty-nine percent of all new podcasts are machine-made, the platforms become less useful, not more. You're wading through noise to find signal. The person who spent weeks researching and recording a genuine episode gets buried under thousands of synthetic alternatives.
So the real problem is volume?
Volume is part of it. But it's also about trust. Once you learn that you can't tell the difference between human and machine, you start to doubt everything. That erodes the whole ecosystem.
What would labeling actually change?
It gives you back your agency. You can choose to listen to human creators if that matters to you. Creators can be recognized for their work. The market can function again instead of collapsing under synthetic weight.
Do you think platforms will actually do this?
Only if they have to. Right now, AI content is cheap to host and generates engagement. There's no financial incentive to label it or restrict it. The pressure has to come from listeners and creators refusing to accept the current state.