From: redpointai
AI has significantly influenced the music industry, with Suno emerging as a viral product, boasting over 10 million users who have generated songs [00:00:11]. Mikey Shelman, CEO of Suno, shares personal experiences and insights into how people interact with and create music using AI.
The Personal Connection to AI-Generated Music
Mikey Shelman highlights that music creation with AI isn’t solely about the final product but also about the journey and the personal connection.
A Favorite Community Creation
Shelman’s personal favorite Suno creation is an Estonian language German lead song [00:01:01]. Despite not speaking Estonian or typically being interested in the genre, the song deeply resonates with him, having listened to it hundreds of times [00:01:17]. He attributes this connection to the user’s care in crafting a vision that strongly resonated with him [00:01:32].
“There was an incredible amount of care taken by the user who made that to really craft it in a way that I guess was the vision that they had and for some reason that Vision resonated very strongly with me for reasons I can’t entirely explain” [00:01:32].
Family Collaborations
Shelman frequently creates music with his son, often about his son in fantastical situations [00:02:06]. One example is a song about his son driving a Zamboni, which three-year-olds are not allowed to do in Massachusetts [00:02:16]. These songs resonate because of the shared experience of crafting music with his son, emphasizing the journey over just the final product [00:02:29]. He views these songs as marking moments in time that he will never get back [00:02:57]. He believes people will revisit these songs in the future, similar to how they look at old photos [00:05:19].
The host also shares a personal anecdote, creating a country song for their producer about the many boxes of podcast equipment sent to their house [00:00:04], highlighting the social and humorous aspect of AI music creation.
The Joy of Collaborative Creation
Shelman believes that making music with other people offers some of the most enjoyable moments in his life [00:02:49]. This joy of the journey is something Suno aims to amplify [00:13:27]. He sees music as a conversation, emphasizing that future development will focus on “multiplayer” experiences, including synchronous jam sessions or asynchronous co-creation where users build on each other’s musical ideas [00:14:30].
Surprising User Behaviors and Product Evolution
Insights gleaned from user interactions have shaped Suno’s development, revealing unexpected ways people engage with the platform.
The “Soundtracking Your Life” Phenomenon
Suno users can be broadly categorized into two groups:
- Casual Users: These users “soundtrack their life,” narrating everyday experiences musically [00:06:13]. Examples include songs about Starbucks getting a name wrong or unexpected package deliveries [00:06:34].
- Power Users: These individuals use Suno as a creative outlet, enjoying both the process and the final product [00:06:50]. They spend hours crafting specific sounds and stories they have in their heads [00:07:05].
[!NOTE] Democratizing Music Creation: A key learning has been that many people possess great musical taste and ideas but lack the means to produce them due to the steep learning curves of traditional music production software like Ableton or Logic [00:07:26]. AI tools like Suno allow for reimagined processes, enabling people to create music in completely different ways [00:07:52].
Unexpected Pride in Creation
A surprising observation was how users reacted when their songs appeared on Suno’s trending page [00:47:58]. They would edit their song titles to include their names, demonstrating a strong sense of pride in their creations [00:48:00]. This confirmed the importance of allowing users to feel proud of their output [00:48:06].
Shifting Platforms
Initially, Suno gained significant traction on Discord, and Mikey Shelman anticipated it would remain a primary platform [00:48:45]. However, upon releasing a “thin, underpowered web app” that lacked some Discord functionalities, 90% of usage shifted to the web within five days [00:49:08]. This experience highlighted that for building an “all-encompassing music experience,” a dedicated web application is more suitable than a messaging platform [00:49:23].
The Role of Community Feedback
Suno benefits from a large and engaged user base, particularly an “amazing Discord community” that provides implicit and explicit feedback [00:22:18]. Users implicitly show preferences by choosing between models or by their usage patterns [00:22:24]. Explicitly, they report issues, providing crucial guidance that prevents the team from “flying blind” [00:22:32].
Vision for the Future of AI Music Creation
Shelman envisions a future where AI music creation is more intuitive, interactive, and integrated into daily life.
Beyond Text Prompts
Currently, most AI creative tools, including Suno, are text-driven, which Shelman sees as a sign of how early the technology is [00:10:35]. The “blank canvas problem” of starting with a text prompt is a significant challenge [00:08:14]. In the future, he hopes for more intuitive interactions beyond text, such as humming a melody, tapping a beat, or using visuals to inspire the model [00:10:01]. He believes this will allow people to truly “pour their heart out” into their creations [00:13:11].
The Playful Aspect of Music
Shelman aims to bring back the “form of play” that adults often lose with music, similar to how young children freely interact with it [00:03:50]. He compares this to “paint by numbers,” where a guided creative experience can still be deeply enjoyable [00:04:25]. The goal is to lower barriers to music creation, allowing more people to enjoy the process and express themselves [00:11:07].
Interactive Live Experiences
Shelman was deeply impressed by a Twitch streamer who used Suno to host a “digital concert,” engaging a large audience who could interact and micro-pay the streamer [00:16:51]. This demonstrated the potential for interactive, group-based experiences that can bring people together through music, much like a live sports event [00:17:21]. He envisions a future where fans and even athletes could provide input to create game-specific songs at stadiums [00:17:51].
Vision for Future Capabilities
While not wishing to reveal specific future capabilities, Shelman notes that the ceiling for AI music quality is “quite a bit higher” than simply producing a 3.5-minute pop song indistinguishable from human-made music [00:37:50]. He desires more iterative control, allowing users to specify exact changes when a song is “wrong” [00:23:55]. Personally, he dreams of a Vision Pro app that allows him to play “air guitar with a band” or conduct a symphony in real-time [00:38:23]. He believes this gamified, intuitive interaction is key to unlocking the full potential of AI music.