Sharing a draft post here before sharing more widely - would welcome comments to help me frame well and refine and identify weak points.
Regarding our recent discussion about Research group sub-circles, the question of the impact of ever-progressing AI to either offer pathways or present obstacles to a Second Renaissance seems like another good subcircle topic.
EDIT 24/4 Replaced Draft post with version posted on substack: I Tried 8 AI Tools to Make Myself Wiser - by Jonah Wilberg
To me, the âwisestâ thing you can do with the draft is just continue your current momentum with it and see how it lands! At the very least, itâs good for setting out the problematic.
That said, Iâve got my own take on what âwisdomâ is in current circumstances. To set some context, my current reading stack includes: 1) Habermas on Aquinasâs appropriation of Aristotle, 2) an analysis of Bahåʟà theology against the background of Sufism and neo-Platonism, and 3) the collected works of Nassim Taleb. That gives a lot of âwisdomâ options to select from! Of course Iâm also already familiar with Vervaeke and everything metamodern. Then, for professional reasons, I listen to multiple YouTubes every day on AIâs impact on careers and the workforce, seeking the âphronesisâ pathway for all that.
Itâs hard to say if my experience generalizes, because as suggested in the context-setting paragraph above, I bring my own content horizon to the party. A lot of my AI use is just to get a second opinion on ideas I already have. (Like using AI to clarify if I properly grasped what Habermas is saying about what Aquinas is saying about the nature of being). Iâm absolutely not using AI to make life decisions for me! Aquinas is reputed to have used 3 or 4 scribes simultaneously to copy his many thoughts. My relationship with AI is like that. (Iâm not in Aquinasâs league as a thinker - few are - but can you picture that guy with Internet access and dozens of AI research assistants!)
OK, so to bottom line it - for me âwisdomâ is nothing about what AI says. Wisdom is about how much information humans can digest and orchestrate at a gut level, and then push back out into the world in the form of prudential action. Vervaekeâs âparticipative knowingâ is an excellent angle on that, but all the comparative philosophy and theology from the world over also factors in. With respect to AI, there is a very, very clear trend taking shape. AI can write code and create artifacts faster than humans can understand what is being produced. The most excellent humans are the ones who bother to understand what is being produced. By âunderstandâ, I mean eat it, sleep with it, live with it. My âwisdomâ move today will be kayaking for a couple hours to let all the ideas work their way into the meat and the bones. No meat and bones, no wisdom.
First, big thank-you for sharing in advance for input.
Any chance of a short SCQ(H) + issue tree? (cf https://issuetrees.com/)
Scanning that would allow me to provide much more helpful quicker input.
And perhaps one for Over the Mountains âŚ
The effectiveness of wisdom-enhancing AI may be inversely proportional to how much it sells itself as such.
The more an AI presents itself as something that will make me wiser, the more it encourages the idea that wisdom can be delivered as a finished product. But wisdom is not like that. It is not just something one receives. It is something one must help bring about.
So the state in which one approaches a learning experience matters enormously. This is especially true when what is at stake is not knowledge or productivity, but self-understanding. How much there is to be gained depends largely on whether the seeker is ready to reflect, to question themselves, and to apply what they encounter.
A âwisdom AIâ and a normal general-purpose AI are, in one sense, just two ways of framing an interaction that might contain something insightful. But the framings are not neutral. âWisdom AIâ suggests a finished product, something refined in advance and ready for consumption. A normal AI leaves more of the burden on the user. It requires more active participation in refining the question, testing the response, and drawing out what matters.
It is like the difference between mining for gold and having gold handed to you. In this case, though, the mining is the point. The digging, sorting, and recognizing are not separate from the acquisition of wisdom. They are the acquisition of wisdom. If wisdom depends on reflection and self-confrontation, then the very idea of wisdom-delivering AI contains a confusion. It turns a process of inner growth into a product category, and in doing so risks defeating the point.
I wonder if this partially explains why explicitly wisdom-branded tools often seem weaker than general models with good prompts. The former are more suggestive that wisdom will come from the system rather than from the userâs own engagement with what the system brings up. The more I believe an AI will make me wiser, the less i believe that I will make me wiser - that the impetus is on me to âbring forth what is within meâ, as it is expressed in the gospel of Thomas saying 70:
âIf you bring forth what is within you, what you bring forth will save you. If you do not bring forth what is within you, what you do not bring forth will destroy you.â
Whatever else one makes of that saying, it captures the structure of the problem. What matters is not just exposure to a teaching, but bringing something forth from within oneself. A good interlocutor, whether human or artificial, does not deliver wisdom as a package. It helps draw out what is latent.
Thanks for the feedback - really enjoying your Habermas-inspired commentary, and looking forward to reading more Habermas myself soon!
As I say in the post, Iâm a big fan of Vervaekeâs model of wisdom, and obviously my take in the post is not intended to be at odds with that model - on the contrary itâs an attempt to show that itâs possible for AI tools to be used in a way that can support wisdom on that kind of model. Your comments made me think that I maybe need to clarify in the post that Iâm not proposing anything like offloading wisdom or decision-making onto AI - itâs about using AI to increase the human capacity for wisdom and decision-making, just as we might have previously approached therapists, life coaches or spiritual teachers.
Thanks - yes SCQ might be helpful in clarifying the connection to the Second Renaissance project I hinted at above:
Situation: In addition to presenting major global risks, the acceleration of AI could potentially offer pathways towards a cultural shift towards wisdom (as part of a Second Renaissance), for example by making processes similar to therapy, coaching and spiritual teaching widely available at low cost.
Complication: It is unclear whether such tools can really take the place of human therapists, coaches etc, and whether it is either theoretically or practically possible to gain the kind of wisdom we need through interaction with an LLM-based AI.
Questions: Can current AI tools make at least one person (e.g. me) wiser, using a rigorous definition of wisdom? Which current approaches work best for this purpose?
The question I plan to explore next - though leave out in this piece - is what mechanisms might support the spread of processes that actually work, as opposed to processes that spread âfake wisdomâ.
Thanks, interesting points, and I agree to an extent. Presumably youâd acknowledge though that some interlocutors are better than others in drawing out what is latent? The existence of therapy, coaching etc as professions depends on that fact. Typically such professionals donât advertise themselves as simply delivering a finished product, and to be fair to the wisdom AI apps, neither do they. But both do advertise an effective process of âbringing forth what is withinâ - and my personal experience tells me this will often involve challenging the clientâs own framings with alternative framings and positive suggestions, rather than simply reflecting back the clientâs own âinnerâ assumptions.
Yes, for sure. Particularly insofar as their method is not to give answers or solutions, but to help the seeker find them on their own.
I donât have a problem with wise teachings being made available via AI. Itâs more the framing that AI can make us wiser, which seems to imply that the AI is active and we are passive. Another way to frame it might be, can AI be helpful for those seeking wisdom? It may seem pedantic to make much of this primarily grammatical difference. But Iâd argue that the difference that makes the difference, for those who become wise versus those who fail to become wise, is the active rather than passive nature of their search.
My response above was my human, analog, a cappella, unplugged version. Below I just asked Gemini point blank - can AI be wise? Generally, I find Gemini aligned with my prior thinking. It also adds some detail and color, which is one thing AI does well. Then, to finish the human-in-the-loop loop, I âownâ what Gemini is saying by making the equivalent of margin notes about its structured output.
Gemini: The question of whether AI can be âwiseâ moves beyond simple data processing into the realm of philosophy, ethics, and cognitive science. To answer this, we have to distinguish between functional wisdom (the ability to give wise advice) and ontological wisdom (the internal state of being wise).
In many ways, AI can simulate wisdom by acting as a high-level synthesizer of human thought. Because it has âreadâ millions of pages of philosophy, literature, and history, it can:
Identify Patterns: AI can recognize systemic issues (like the âmetacrisisâ or complex adaptive systems) that a single human might miss.
Provide Perspective: It can offer multiple viewpoints on a single issue, which is a hallmark of wise judgment.
Lack of Ego: Unlike humans, AI doesnât have personal insecurities or biases rooted in self-preservation, allowing it to provide objective, âcool-headedâ counsel.
Despite its ability to sound profound, AI lacks several core components traditionally associated with wisdom:
| Component | Human Wisdom | AI Simulation |
|---|---|---|
| Lived Experience | Built through suffering, joy, and physical interaction with the world. | Derived from statistical patterns in text. |
| Empathy | A biological resonance and shared feeling with others. | A mathematical prediction of what an empathetic response looks like. |
| Moral Agency | The ability to take responsibility for the consequences of a decision. | No âskin in the gameâ; the user bears all responsibility. |
| Contextual Nuance | Understanding the âunspokenâ rules of a specific culture or moment. | Relying on the data it was trained on, which may be outdated. |
There is a vital distinction between intelligence (the ability to solve a problem) and wisdom (the ability to decide which problems are worth solving).
Intelligence is knowing how to build a powerful autonomous system.
Wisdom is asking whether that system should be built at all, considering the long-term ethical governance and impact on humanity.
The Verdict: AI is an incredible âwisdom mirror.â It reflects the collective insights of humanity back at us. It can help us navigate complex models or architect moral frameworks for agentic systems, but it doesnât âpossessâ wisdom because it doesnât have a life to live or a soul to weigh. It provides the map, but you are still the navigator.
Do you think a âmoral compassâ for an AI can ever be more than just a set of programmed constraints?
Human commentary:
+1 on the pattern recognition part.
On lived experience, Vervaeke of course. But also BahĂĄâuâllĂĄh, using language of the Sufis - the pathway to unity lies through the valley of love, and without pain, there is no love.
On moral agency, âSkin in the Gameâ is literally a Taleb title. Taleb, in a nutshell, expounds about why pattern recognition is not enough. Although Taleb is about as secular as it gets, I feel like he would vibe with BahĂĄâuâllĂĄhâs perspective about no pain, no love. In Talebâs terms, offloading important decisions to AI is for âsuckersâ.
+1 again on empathy and contextual nuance.
Finally on Habermas and Aquinas - Habermas is in search of the birth of the modern, pure and simple. Through a lifetime of reading everything that matters in history of philosophy, he knows full well that the modern hatches out of High Scholasticism, which in turn results from the importation of Aristotle (through Islamic transmission) on top of the Christian neo-Platonism of Augustine. The theological problem space of Aquinas also fills the same dimensions as that of BahĂĄâuâllĂĄh and the Sufis - what is the relationship of a God who saves and sets moral expectations to the God who creates the world in the first place? Or more generally, how does moral order relate to natural order - how does history fit into cosmos?
Great stuff. Yea Iâm generally sympathetic with the âAI canât really be wiseâ and even the âAI isnât real intelligenceâ camp for the reasons you cite. Thatâs why I tend to clearly distinguish âcan AI be wiseâ from âcan AI make us wiserâ. One might conclude that using AI for wisdom is more similar to divination techniques like I-Ching - which purport to draw wisdom from complex patterns in unintelligent objects like yarrow stalks, and I think Iâm OK with that analogy!
Circling back to your draft Google doc paper above, itâs clear that you specify âAI to make us wiserâ rather than AI being wise itself. So itâs a tool. Weâre agreed on that. Can the development of wisdom avail itself of tools? Likely yes, although my reading of various mystical traditions suggests there comes a moment to leave all such devices far, far behind. Beyond that, any old tool will do. Or no tool. Whatever.
What seems on offer through most of these AI wisdom-generators is self-awareness. They challenge us to examine and justify whatever we assert. To me, that makes a sort of sense in developmental context - a person who has largely led an unexamined life (like 19 year old me, back in that day) will surely benefit from personal reflection in response to prompts. After the âselvingâ process as been cooking along quite a good long while, however, Iâm not really sure why one would want to responding to chat prompts furnished by companies whose ethical credentials are themselves far from beyond reproach. Chat bots do not embody wisdom. I really doubt they will ever be capable of this: What is Darshan and How to Receive it
At the recent UTOK conference, Rob Scott offered a basic stage theory of selving like this:
Level 1 - Unconscious Selving
Level 2 - Managed Selving
Level 3 - Conscious Selving
Level 4 - Fully Awake Selving
My guess is Chat bot wisdom coaching might nudge a level 1 to level 2, or a level 2 to level 3. At level 3 itâs time to step away from the keyboard and face experience in the raw.
Yes good points, and Iâve amended the phrasing in the version I posted to substack I Tried 8 AI Tools to Make Myself Wiser - by Jonah Wilberg
I still think thereâs a sense in which âAI makes us wiserâ is useful, in that most of us are quite passive with regard to the inexorable development of AI tools, and the question is whether this technological shift can have positive impacts (as well as the more obvious negative ones) for wisdom.
But overall I agree, itâs important to be clear on the fact that becoming wiser with AI tools has to be a highly active process for the user.
Yes I did wonder whether something like that stage theory might be a way of understanding the effectiveness of AI tools. My current thinking is more along the lines of the multidimensional stage theory of late Wilber - if there are multiple dimensions of development: growing up, showing up, waking up, cleaning up and maybe just cognitive complexity, which of these can be supported by AI? My sense is quite a few, even if non-AI practices need to be part of the mix too.
At this point, AI is indeed part of my personal post-capitialst, volunteerist, prosocial, (2R?) praxis, so consider what follows an experiential data point, not as propositional arguments asserted for generalized application. Mileage will vary!
I got into AI from the Ikigai angle of âwhat does the world need?â. More precisely, âwhat do my CS students need?â, because industry advisors made it crystal clear in recent months that CS students without AI skills are functionally unemployable. Working back from there, to teach AI it seemed necessary to learn to use AI, so experiments thus began. âWhat do I love?â Not burning endless fossil fuels and paving over rural landscapes with AI data centers! From a personal integrity POV, I am the exact opposite of a âtoken maxerâ. Iâm trying to be economical with AI like I am with hot showers or airline flights. I believe the day is coming when the free tier of AI will be rather limited (to reflect its true cost) and most of us will largely be using it for search (assuming the grid itself does not collapse). That said, I used AI to run some numbers on AI session costs, and the sorts of queries I run may burn on average $0.01-$0.05 of underlying water, electricity, physical capital, etc. Compare that to driving 10 miles in a truck to the local university library. AI gets more targeted results at lower net environmental cost. (The library does still merit the occasional visit, however!)
So in Wilberian terms, my point of entry to the AI world was in âshowing upâ. Do you think you can untangle with metacrisis without AI? Then please do! My own sense of karmic social obligation however is I am duty bound to seize the reins of AI and ride it like a cavalry officer. With aplomb. Having thus put AI through its paces, on the rest of the Wilberian spectrum, Iâd say the sweet spot is largely âgrowing upâ, which essentially equates to cognitive-emotional development. On the cognitive side of that, for me, AI is part of a current quadrivium of purchasing physical books, YouTube, social media, and targeted AI queries for follow up on the other three. All that is making me wicked well-informed and transdisciplinary AF. However, as St. Paul so aptly put it, âknowledge passeth awayâ, so moving on to other targets ⌠Does my sense of self or ego require AI? Only because of current historical situation. AI is not pushing me to higher Kegan stages. I only use AI at all because of the Kegan stage I have already arrived at. If we start talking nth tier beyond that, the self becomes more of a mildly amused spectator in the field of cognition, not something that needs a lot of stocastically calculated digital word-smithing for anything that really matters. âCleaning up?â. No. My Jungian shadow manifests in dreams and difficult emotions. I eat those raw. AI for that would just be a distraction from the task at hand. âWaking up?â. Not really. Yes, itâs possible to use AI to gain insight into the expressive productions of the great spiritual teachers of all times and all ages, and I do that all the time. But seriously, thatâs like reading Cliffâs Notes on Hamlet as opposed to attending Hamlet in live performance, let alone acting in the play itself! I canât think of a single spiritual teacher who would say reading AI summaries of spiritual teaching is anything beyond the bare novice level of the search.
To sum up, in 2026, AI is basically Arjunaâs battle chariot, so those karmic warriors among us need to armor up and lay in to the fray. For those who are more focused on the âgrounded in beingâ part that comes prior to accepting such sacred action-duty, Iâd say retreat to the forest will generally be more productive.
We seem to be staggering under the sheer weight of text, already heavy in the pre AI age, now we are AI turbocharged, technologically enabled to quote, connect and citation orgy ourselves into rarified complexity orgasmic petit morte.
Robertâs âno meat, no wisdomâ is meaty wisdom.
My personal experience, is that sparring with AI, has been transformative via negativa, finding the strength and power to hold my humanity, my contemplative practice, my consequential self. I have found what is humanly beautiful and dear and fragile, what can be lost, by seeing its lack in AI.
AI is all expression, and no felt sense. Iâm increasing my intimacy, and my felt contact with life, whilst assisted in my expression by an AI. Not exactly a paradox, but close.
One of Jane Austinâs characters remarked that visitors always bring pleasure, if not in the coming, then in the going.
Likewise AI seems most valuable in revealing the negative space it creates.
Yesterday I found myself in a face-to-face meeting of this group (with some of the same people pictured on the web site) and in one of the many conversation topics AI came up. AI is a tool I have learned to use (and not use) depending on what the job is. But as I shared at the meeting, itâs sure as heck not my therapist, by boy friend/girl friend, confidante, spiritual director, or personal bestie. Tool. Full stop. But given that I teach CS, itâs a tool I need to be decently good at using, to teach students who are moving into a jobs environment in which skill with that particular tool is about all anyone much cares about for now.
So the question came up, how to avoid the siren song of AI? My simple minded answer - in a physical room full of physical people having physical chat munching physical snacks - was get off the keyboard for awhile! Just be a well-rounded person. AI will find its place. AI is overblown in direct proportion to the degree the remainder of our lives is underblown.