When comparing traditional guided meditation apps, like Headspace and Calm, with modern AI meditation apps, there are obvious upsides and downsides.
At their core, both offer guided meditation sessions that are suitable for beginners through to advanced users. So why use one over the other?
The benefits of AI meditation
AI meditation offers another level of personalisation. Instead of generic lessons, the teachings are specifically chosen to fit in with the listener’s current situation. Let’s illustrate this with an example.
You’re on the train, heading into the city to meet up with some old friends. You haven’t seen them in a while, so you’re feeling excited but a little anxious.
Using a traditional meditation app, you may look to seek out a specific session that helps to reframe these emotions. Maybe one on excitement? Or anxiety? But this doesn’t really fit into either. What you’re feeling is a unique mix of both.
You also don’t want to start a meditation anxiety course just for this one time. You’re looking for something for the present moment, not for a long term course.
Instead, you’ll have to pick a generic session, and hope that the teaching is somewhat relevant. You do this, but quickly realise that the techniques being used in the practice are also not suitable.
What if you don’t want to close your eyes? Why is it telling me to listen to the sounds of my house when I’m on a train? This is hardly an immersive experience.
Enter AI meditation.
Through a simple casual chat, all of these details can be input as context for the session. The specific emotion you’re feeling becomes the target of the teaching, and the instructions during the practice actually match with your environment.
Assuming high quality audio and appropriate amounts of silence, this is a flat out win for AI meditation.
So what’s the problem?
The weakness of traditional meditation apps serving rigid sessions is also its greatest strength. This rigidity ensures consistency and structure.
Users of the apps can be sure that when they’re taking the beginners meditation course, they’re learning the same things that everyone else is learning. You trust that the teachings and techniques used have been developed and checked by experts.
AI meditation doesn’t have this. Sessions are generated, and are not backed by specific teachings or written by expert teachers.
Their ability to be creative and adapt the meditation to fit with the listener also means that they cannot be fully trusted to deliver actual teachings and techniques. How can you be sure that what you’re being taught is actual meditation literature?
With most AI meditation apps you cannot. That level of trust isn’t there, and for beginners looking for expertise, AI meditation doesn’t offer any guarantees.
If you cannot trust your teacher then you will never be able to learn. Hallucinated teachings are of no benefit to anyone.
Our solution
This is an obvious and significant barrier for all AI meditation systems, and so solving this problem has been a major priority for InTheMoment.
The first problem to solve was consistency. How can we ensure that each session builds upon the last, gradually introducing new teachings and techniques, without repeating or jumping?
We achieve this by making sure that each new session has a full understanding of the user’s previous sessions. It knows which teachings were explored, how they were introduced / described, and the techniques used in the practice.
Any session feedback that the user left is also presented, allowing subsequent sessions to take into account user preferences. Did a particular technique resonate strongly? Was a certain explanation weak or confusing? All of these pieces of context are used in the construction of every session.
Fixing the problem of structure is a harder challenge. We already described how AI meditation’s biggest benefit is in how each session can be perfectly adapted to the user. So how do we add structure to this without damaging the amount of personalisation that can be achieved?
InTheMoment uses playlists, which can be thought of like structured curriculums. When following a playlist, the exact teachings and practices are human defined, similar to how a teacher uses a curriculum to know what to teach.
The AI then personalises the exact delivery of the session, allowing the sessions to still stay relevant to the user. This approach ensures that the content within each lesson remains consistent for all users, adding a level of trust which is usually not there in other AI meditation systems.
The playlists themselves contain content sourced from books, online resources, and other meditation literature. They are all human written and based on a collection of teachings relevant to the topic of the playlist.