I asked AI for meditation that fits my view

I also asked an AI to critique the view itself. This page is the other half of that day: I asked the same assistant which meditation practices would fit the view, if the view is taken seriously on the cushion.

The model was Grok Bot, an AI assistant, run on 6 November 2026. What it suggested is not an endorsement, and it is not instruction from me. I am a practitioner, not a teacher. I do not claim attainments. Nothing here is medical advice. I am interested in fit with a view of consciousness, not in selling a method.

The view I handed over

The short version of my view that AI is not conscious is the same one I live with at work and on the cushion. Consciousness, as I use the word, means there is something it is like to be the thing. Clever is different. In the agents I build I cannot find action nobody prompted, or anyone there to feel what happens.

Consciousness seems to me as fundamental as gravity. On the cushion, the personal "me" can fall away and awareness does not. That is the reading I take from in the seen, only the seen: no you seer in the sense the sutta means, and still something aware. I cannot fully square those, and I hold both for now.

My favourite practice is anapanasati, mindfulness of breathing, as laid out in the Anapanasati Sutta. I have written about that in Who comes back to the breath?. I care about jhana as the suttas and teachers I read describe it. I do not teach it.

That package, plus the Machine and mind essays, is what I handed to the model. I asked for practices that fit the view. No health claims. No new doctrine.

Which practices did the AI say fit this view?

Grok Bot suggested seven. I paraphrase the list and keep its "why it fits" notes short. The steps below are the model's suggestions, attributed as such. Where I already publish a version, I point there instead of retelling it.

1. One-minute Bahiya on sound. The model said this practice sits closest to the cut I care about: raw contact before the self-story. Set a timer for one minute. Wait for a sound. Let it be heard. When a label arrives, notice that the label came second, and return to the sound. I already published a version of this in the Bahiya essay, so the model was mostly handing my own practice back to me.

2. Anapanasati, first four steps only. Knowing the breath as heavy or light, experiencing the whole body, stilling the physical process. The model kept to the first tetrad and pointed at the nostrils and upper lip, which is how I already sit. It did not retell all sixteen steps. Good. That work belongs on the anapanasati essay.

3. The drift button. Sit with the breath. Each time you notice you have wandered, mark it once, then return. Do not hunt for drift. Wait until noticing happens. The model tied this to the twin-sit thought experiment in the anapanasati essay: who notices, and who comes back, without being prompted. That is the same agency gap I keep finding in machines.

4. In the thought, only the thought. When a thought appears, mark it lightly as "thought", not as its content. Return when you notice you have been inside the story. The model linked this to Sujato's rendering, which puts thought on Bahiya's list, and to the way I ask agents to keep text apart from the person felt behind it. Related to noting and labelling, without turning this page into a noting manual.

5. Whole-body breath. After the nostrils are clear, widen awareness to the whole body with the breath, toward the bathman simile in MN 119. Do not invent sensations. Do not chase a nimitta. The model noted that Ajahn Brahm's map and Leigh Brasington's map disagree on entry, which I already say in the anapanasati essay.

6. Peg on the story. Name three surface facts of a moment already happening. Notice the next sentence the mind offers about you. Return to the facts, then act from them as needed. The model cast this as a cousin of feel versus detect: keep the information, drop the added story. The cut is the practice.

7. Between breaths. Know the small pause after the in-breath and after the out-breath without lengthening it on purpose. The model tied this to a line of Ajahn Brahm's I keep thinking about, and to my point that an electronic nose has samples and no "meanwhile". The pause is a between for someone.

What already fits: anapanasati

The model's second suggestion pointed straight back at the practice I already treat as home. That is the right move for this page: support the live essay, do not replace it.

Why does anapanasati stay my favourite?

Simplicity. The Buddha's own instructions, as far as two good translations can tell me. And one object: the breath. A lot of meditation advice asks you to listen to music, imagine a field of flowers, picture a light. The Anapanasati Sutta gives you the breath. One thing, and it is the one I keep coming back to.

I rest attention at the nostrils and the upper lip. The upper lip takes time. Ajahn Brahm prefers not locating the breath; I do, and I say so. I refer to his map for jhana, including the nimitta. I think, though I am not sure, that I once experienced a nimitta. I leave it at that.

The sixteen steps are already on the site. This page will not re-run them. If you want the table in two translations, the twin sit, and the question of who comes back, read Who comes back to the breath?.

What I would try, and what I would skip

The model's list is a menu, not a vow. Here is how I filter it, as someone who already has a favourite practice and a view to protect from decoration.

What would I skip, and why?

What I would keep close: the first four steps of anapanasati, the one-minute Bahiya on sound, the drift button, and the pause between breaths. Those sit inside practices and cuts I already write about. They fit the view without adding anything extra.

What I would try gently: whole-body breath, as a widening of step 3, without turning it into a hunt for signs. In the thought, only the thought, as a short drill next to Bahiya, especially on work days full of labels.

What I would handle with care: Peg on the story. The cut between facts and story is real for me. I do not want it turned into a technique for managing a life. The model kept it free of promised outcomes. I would keep it that way, and I would not drag it into the recovery pillar from here. Craving and phones have their own home under Recovery and attention if I ever write that out as practice.

What I would skip: anything that invents new objects, visualisations, music, or app-guided sessions. The model stayed off those, which matched the brief. I would also skip any suggestion framed as a ladder to attainment. That is not how I write, and it is not what I asked for.

Vipassana versus anapanasati did not need a contest here. The model stayed with breath, Bahiya, and noting-like marks as servants of those, not as a rival school.

What this is not

This is not teaching. I am not a teacher.

This is not medical advice, and it is not a treatment for any condition. Meditation is practice, not treatment. For diagnosis or support, talk to your GP or a clinician.

This is not a benefits list. I asked which practices fit a view of consciousness. Fit is not a result. I am not claiming any outcome for any of them.

This is not an app review, and it is not a guided-meditation catalogue. Stay with the practices and the view.

This is not a change to the stance. AI is still not sentient, on my view. A model can sort practices by fit and still have nobody home to sit them.

More in Practice. For the sutta side, see The seen. For the view these practices were matched to, see Machine and mind and the critique I asked for on the same day.

FAQ

Which practices did the AI say fit this view?

Seven: a one-minute Bahiya on sound, the first four steps of anapanasati, a drift-and-return sit, thought as merely thought, whole-body breath, a Peg-style cut between facts and story, and knowing the pause between breaths.

Why does anapanasati stay your favourite?

Simplicity, the Buddha's own instructions, and one object. Details and the sixteen steps are in Who comes back to the breath?.

Are you teaching these practices?

No. I am a practitioner quoting an AI's suggestions against a view I have already published. I do not claim attainments.

Does this page claim benefits?

No. It is about fit with a view of consciousness, not about outcomes.

How this was made: the view, the filter and the final edit are mine. The practice list is from Grok Bot, an AI assistant, run on 6 November 2026. It used my published essays and did not invent experiences or quotes of mine.