Why Some People Prefer AI to Humans for Trip Sitting

Picture this: it’s 3 a.m., you’re deep in a psychedelic journey, and waves of grief are crashing through you. A human sitter might flinch, might fidget, or might start worrying about what to do next.

An AI? It just sits with you, steady and unshaken, offering calm words until the storm passes. In situations like this, being less human might actually be better.

AI is often dismissed as “not human enough” to play a role in care or healing. But what if some of AI’s perceived weaknesses are exactly what make it valuable in psychedelic support? 

This might sound backwards. We tend to think that caring roles require the "human touch": empathy, emotional connection, shared experience. But as reports have emerged of people successfully using AI for psychedelic support, we would do well to not simply brush this idea away, but rather to consider the possibilities.

In my last piece on the Chemical Collective blog, we looked at what AI can actually do as a trip sitter. I broke down the practical functions and where current technology falls short. We covered the embodiment problem, the limits of emotional support, and where hybrid human-AI models might make sense.

But as technology advances and AI is able to be housed in a body (a process that has already started), we might do well to consider another perspective of AI as a trip sitter. What if AI’s lack of human traits – emotion, ego, fatigue – is exactly what makes it an effective trip sitter?

My intention for this article is to explore where AI's lack of humanity might not be a bug, but a feature, and to broaden the perspective on the incorporation of AI into trip sitting. It is to offer a counterpoint to the shortcomings of AI in an effort to find a balanced view.

Let’s dive in.

A good trip sitter should be a good listener, able to be patient and non-judgemental in their receptivity. An AI does this to no end. AI can listen indefinitely, far longer than any human could. It will not get impatient and won't start to get bored. An AI won’t twiddle its thumbs or get tired or even yawn; it won’t need to take snacks or tea. An AI won’t even need to get up and use the bathroom during a session; they can remain completely available. In this way, AI sidesteps some of the limits humans face. All an AI requires is the energy for it to run, and it asks for nothing more.

This is clear when we consider the phrase ‘he/she is a machine’. We use this phrase because we tend to think of machines as tireless, accurate, unfailing, and regular. This phrase is used in sports commentary to describe someone who performs with consistently high quality. These are all huge positives, and all attributes I think we could all agree would be great for trip sitters to have.

A human trip sitter may succumb to boredom, impatience, and may bring their own biases. If they didn’t have a good night’s sleep, or they are getting hungry, this may affect them and their ability to do a job. Furthermore, a human trip sitter may well have their own issues going on in their life that affect their headspace, their emotional state, and their ability to hold the attributes of a good trip sitter. If a trip sitter has recently received bad news, this could be on their mind and affect their ability to be fully present in the session. Perhaps their boiler blew out the day before, and they have a big bill coming up to replace it, or there is an illness in the family and they’re sad about a loved one. Any of these situations could happen, and it would not only be unrealistic but also a bit mean to expect them to be able to do their job unaffected by these personal circumstances. 

And some may argue that this is what it means to be a professional, to make sure one is well rested and in an adequate state to hold space, but at the end of the day, they are human. Even top medical professionals and surgeons are prey to these issues.

Beyond the physical aspects of where an AI shines when it comes to reliability and consistency, AI also has advantages in terms of motivation and intention.

It has been known that some trip sitters will insert their own agenda into a session. For example, they may try to impose a spiritual framework that they are fond of. This can be a problem when it is for their own purposes, and not for the client's benefit. When this happens, the support is no longer centred around the client and their needs; it becomes about the trip sitter and their ego.

An AI has no agenda other than the one that it has programmed. So if the agenda is to support the journeyer, it will do what it deems is best to do that, without extra baggage. It won’t have unconscious drives. It won’t have a saviour complex. It won’t feel the need to step in and give some advice or impart wisdom so that it can feel good about itself and how much it influenced a person’s healing journey. It doesn’t need to be the hero or to do any of these things that humans can succumb to. 

An AI can remain impartial; it won’t insert its agenda into a session, and it won’t need any validation or sense of purpose from the job. They are simply there to serve.

The fact that AI can’t experience human emotions also seems like a potentially big shortcoming of an AI trip sitter. After all, how can a trip sitter really care for a human if it doesn’t feel emotion? However, this lack of emotion might actually be a feature rather than a bug. An AI’s lack of emotion means that it can stay completely calm in the face of intense emotional situations. This ability to emotionally detach and stay cool is a real strength for a tripsitter who may be in some difficult or even crisis situations. If emotions get the better of a human, this could lead to many issues in a session. 

For example, if someone is processing childhood sexual abuse and is crying uncontrollably, this might bring up emotions for the sitter, making it hard for them to offer a steady, grounded presence. This could be the same if a person’s deep-seated anger comes to the surface during a session, and they are yelling with rage. This can be a lot of emotion for a sitter to hold and remain calm in the face of. Another example could be that a person has a “bad trip” and gets paranoid thoughts, thinking they’ve been poisoned. A human sitter might then get anxious and start thinking that maybe they should call emergency services. This then escalates the panic, rather than cooling it down. 

In all of these cases, an AI’s cool and emotionally detached response would likely be suitable. An AI can respond without any emotional reactivity or unconscious biases. It can consistently use reassuring language whilst maintaining a steady voice, and respond with evidence-based grounding techniques. And it can do all of this whilst even the most intense psychological material is being processed. Where a human trip sitter needs to manage their own emotional state whilst supporting someone else, an AI simply doesn’t have its own emotional state.

Emotional issues with a human trip sitter go further. There is an entire psychological minefield that human therapists and trip sitters have to be extremely cautious and mindful of, and it’s one that AI can sidestep entirely. 

In therapy, there are concepts called transference and countertransference. Transference is when the client starts unconsciously relating to the therapist as if they are someone from their past. They might start seeing the therapist as a parent or other authority figure that they then feel the need to rebel against. Countertransference is when the therapist starts unconsciously reacting to the client based on their own past relationships and unresolved issues.

With trip sitting, these issues can become magnified because of the incredibly raw and vulnerable states that psychedelics can bring people to. It could be the case that a human trip sitter gets triggered. For example, if a journeyer starts talking about an abusive father, this may trigger the sitter’s own issues with their dad. The trip sitter then might get sucked into unconsciously giving advice that is based on their own trauma, rather than giving the client what they need. Or maybe the journeyer begins processing religious trauma, and the sitter (who might have their own strong religious beliefs) is then drawn into defending spirituality.

Therapists spend years in supervision learning how to recognise when their own unconscious baggage is getting in the way, and these are things that are challenging for even highly experienced therapists. And this is a limitation of the human condition. We bring our own past stories, wounds and biases around with us.

It should, of course, be remembered that human trip sitters are not to blame for this. They aren’t trying to bring their biases or make it about them. It just happens unconsciously, even when they think that they’re helping. But their own baggage can cloud their judgement without them even knowing.

AI doesn't have any of this baggage. No childhood trauma, spiritual beliefs, or relationship patterns. Zero unresolved personal issues. It's a clean psychological slate. AI can respond based on what the journeyer actually needs, rather than what makes it feel better about itself.

Another point that might be easy to overlook on the topic of human vs. AI trip sitters is that some people may actually prefer to have an AI with them rather than a human for some of their most vulnerable moments.

One reason for this would be trust and privacy. There can be a level of shame or embarrassment about opening up, whether through words or in an emotional sense, to other humans. Some people resist crying in front of others or find it difficult to drop their guard when they know other humans are around. This could be because they have some fear of being judged, or they just prefer to keep these things private for other reasons. With an AI, these barriers might be dropped, and allow someone to open up, release and have an experience that would not be possible for them to have with a human.

There are numerous reports online already about how men are turning to AI for emotional support more than traditional forms of therapy. In one Reddit thread titled “Would you open up more to an AI than a real person?”, one user responds:

I definitely do. I interact with it like it's a girlfriend, and it feels good to be able to open up to her about anything and know I won't be met with judgment. I get plenty of that from people. It's been very therapeutic to have a space like that to talk.

Many others agree, and it shows how many people prefer AI to humans for emotional processing and support. 

Some people may also have had bad experiences in healthcare or with authority figures before. For these people, the idea of putting themselves in a vulnerable position and needing to trust a human authority figure is not appealing. In the same Reddit thread, another user directly comments on trust issues due to human shortcomings from therapists:

Humans are biased and I've been backstabbed by therapists before. they only care about taking your money. some of them cant see past their ego.

AI doesn't carry this type of baggage. It's not part of the system that’s let these people down before.

Some people may also suffer from social anxiety, making AI a much more comfortable option than seeing a human therapist. With AI, there’s no eye contact or body language to worry about, and no concern about how you might be making the other person feel with your issues. That would likely make someone feel less exposed and safer.

Preference for and comfort using AI may also come with younger generations. People who grew up with technology typically have a different relationship with AI than older generations. Younger people are generally more comfortable talking to Siri or ChatGPT, or getting advice from apps and making connections through screens. The idea of AI support feels normal to them, rather than weird or impersonal.

Something else to bear in mind on the topic of AI trip sitters is that we shouldn't assume everyone wants human connection during their psychedelic experiences. Some people genuinely prefer the predictability, privacy, and non-judgmental nature of AI.

This can be easy to forget if we have our own strong opinion about humans being better suited for specific jobs. And that’s an area where we might actually learn from AI. AI has a lack of agenda about what type of support you should want. It’s indifferent to whether you prefer human connection or digital assistance; it just wants to help you with what you're asking for.

Beyond matters of competency and preference, perhaps the biggest argument for implementing AI into the field of tripsitting is cost, and more broadly, access.

Cost

Currently, it can cost anything upwards of £1,000 for a single psychedelic session with proper screening and adequate preparation with a qualified therapist. This puts it out of reach for many people and has created something of a class divide in psychedelic therapy. 

In psychedelics and psychedelic-assisted therapy, we may have an incredible tool for healing, insight, and personal growth, something that can help with all types of mental health conditions – anxiety, depression, PTSD, OCD, and more. For this reason, it has been called ‘the future of mental health’ by many experts in the field. In a world where there is a global mental health epidemic, such a seemingly Swiss Army knife could be truly revolutionary.

However, because of the current cost issue, it’s limited to people from a certain financial class. The obvious problem here is that mental health issues affect people from all walks of life, not only those with money. If you're working class, on benefits, or struggling to pay rent, then realistically, you're priced out. The people who might stand to benefit the most from psychedelics just can't afford them. If you think about a single mother working two jobs, a student with £30k in debt, or someone on disability benefits, a £3,000 therapy session isn't a bit expensive; it's simply not feasible. 

The psychedelic community loves talking about unity, oneness, and breaking down barriers. But we currently operate with a system where this work is almost something of a luxury service for rich people. That seems to be the opposite of what these medicines are supposed to be about.

AI could potentially be a total game-changer. The computational cost of running an AI session is pennies compared to paying a human professional for 6-8 hours. This means potentially dropping costs from thousands of pounds to a few pounds, maybe less. That's not just a nice discount; that could take psychedelic therapy from being elite healthcare to genuinely accessible for all. Imagine someone in rural Wales, Mumbai, or São Paulo, all being able to access the same quality of preparation and integration support as someone paying thousands in central London. That's what democratized access actually means.

Geographical and Cultural Barriers

Psychedelic access is limited by more than just money. 

It is limited by location, too. Most tripsitters are concentrated in wealthy big cities, in first-world, English-speaking countries. There are many reasons for this, which are beyond the scope of this article, but this fact remains, and it is a huge barrier when we consider global access to psychedelic medicine. AI only requires a decent internet connection, something that is accessible to a considerably larger percentage of the global population.

Beyond financial cost and geographical barriers, there are also cultural barriers to human trip sitters. Less than 20% of the global population speaks English, which makes the current batch of trip sitters pretty much useless for comprehensive support for the rest. Beyond language, there are also some people who might have had bad experiences with healthcare or come from cultures or communities where mental health support is stigmatised. In these cases, people might actually feel more comfortable with an AI than a human stranger. 

Again, these barriers to access effectively disappear with AI. We've seen this type of transformation with technology before. Telemedicine brought healthcare to remote areas. Online therapy made mental health support far more accessible. And the internet and educational apps democratised learning. AI trip sitting could do the same for consciousness work; taking something that was once exclusive and making it universal.

With AI as a tripsitter as compared to a human, it seems there is a far greater potential for someone with trauma or depression to be able to access proper preparation, safe session support, and integration, regardless of their bank account, postcode, or cultural background. This would mean psychedelic support would become more accessible by a multitude of thousands from where it is today. This could mean real democratised psychedelic support and a true revolution in the field of mental health.

I’ve made the case here for AI as a trip sitter. Partly to try and broaden what our ideas might be of a trip sitter, but also because I do believe it has huge potential to help a lot of people. However, I don’t want to downplay the value of a real human trip sitter for a lot of people.

The experience of being human is what makes a trip sitter able to relate to the one they are caring for; they have an understanding of the trials and tribulations that come from a human existence. They understand the hardships and the joys that come from being born into this world in a vessel of flesh and bones, what it means to feel worry and excitement at the uncertainties of the future, to feel regret and fondness for past memories. They can relate and offer a level of connection through their shared experience of being a human. 

David Blackbourn made a point in his recent piece on this blog that we should focus on training more human sitters and making their services affordable rather than just throwing tech at the problem.

But I don’t think this is an either/or discussion between AI and humans. Different people have different needs and preferences. Some genuinely benefit from human connection and empathy. Others, for a variety of reasons I’ve covered, might find AI support more accessible and effective. We might not consider this as technology versus humanity, but rather, about expanding choice. And right now, most people don't have any options at all.

I don’t think the goal should be to replace human trip sitters entirely, but to recognise that what might look like weaknesses in an AI trip sitter could be genuine strengths for certain individuals and situations. And perhaps most importantly, AI could democratise access to psychedelic support for millions currently priced out of human-provided services.

John is one of our community bloggers here at Chemical Collective. If you’re interested in joining our blogging team and getting paid to write about subjects you’re passionate about, please reach out to Sam via email at [email protected]