In today’s column, I identify how you can use generative AI and large language models (LLMs) as a vital tool for becoming a deep thinker. The notion is straightforward. You would regularly engage in conversations with generative AI to significantly aid and deepen your thinking processes. This might be likened to the Socratic method whereby a weighty and ongoing dialogue with an astute interlocutor or partner can demonstrably improve your mind.
Let’s talk about it.
This analysis of an innovative AI breakthrough is part of my ongoing Forbes column coverage on the latest in AI including identifying and explaining various impactful AI complexities (see the link here). In addition, for my comprehensive analysis of how AI is being used specifically in medical and healthcare applications, such as mental fitness, mental health guidance, overall cognitive therapy, and mind-improving care, see the link here.
Deep Thinking As Deep Thinkers
Socrates in approximately the second half of the 5th century BC devised a back-and-forth or Q&A process as an alternative method of learning about the world. This became known as the Socratic method and remarkably has stood the test of time. It is extraordinarily simple. Two people aim to jointly and with intensity deliberate with each other, testing each other, refuting each other at times, and otherwise pursuing an in-depth cross-examining dialogue.
This seemingly innocuous but remarkable approach still works nowadays. Why so? Because humans are still humans. We fortunately continue to be conscious thinking beings. The deliberations between erstwhile-thinking people can be a tremendous boon to mental advancement for either or both parties.
The key now is that you can employ the same method nearly freely and endlessly with the use of modern-era generative AI and LLMs. To clarify, this does not somehow anoint contemporary AI with having sentience. We do not have sentient AI. I realize that might be a shocking statement since we have daily blaring headlines that claim otherwise. Don’t fall for the headline blarney.
Generative AI can impressively simulate dialoguing somewhat on par with that of sparring with a human, see my explanation of how the AI computationally and mathematically does this, at the link here. It isn’t the same as verbally scuffling with a human, certainly, but it comes close enough that you can leverage the AI to your advantage as an avid conversational partner or interlocutor.
Human-AI collaboration begets a contemporary Socratic method.
Example Of Using Generative AI For Deep Thinking
Most of the major generative AI apps are free to use or available at a low cost. You can use them anywhere and at any time, 24×7. No need to schedule time to meet with the AI. Just log in and carry on with a conversation.
I am assuming that you would like to see an example of how generative AI comes into this realm. I went ahead and logged into ChatGPT by OpenAI to come up with an example for you. You might find of keen interest that ChatGPT garners a whopping 300 million weekly active users. That is a staggering amount of usage.
I briefly conducted an additional cursory analysis via other major generative AI apps, such as Anthropic Claude, Google Gemini, Microsoft Copilot, and Meta Llama, and found their answers to be about the same as that of ChatGPT. I’ll focus on ChatGPT but note that the other AI apps generated roughly similar responses.
Let’s get underway.
- My entered prompt: “I am aiming to become a deep thinker. For this conversation with you, I want to understand more about happiness. Can you help me think deeply about it?”
- Generative AI response: “Certainly! Let’s start with a question: What does happiness mean to you? Is it a fleeting emotion, a state of being, or something else entirely?”
- My entered prompt: “I think it’s a state of being. It’s when everything feels aligned and good in life.”
- Generative AI response: “Interesting. If happiness is a state of being, do you think it’s entirely internal, or is it influenced by external factors like relationships, achievements, or circumstances?”
- My entered prompt: “It’s probably both. Internal mindset plays a role, but external events definitely affect it.”
You can discern from the dialogue that the AI immediately went into a kind of Socratic method mode.
I bring up as a tip for you, namely, make sure to explicitly tell the AI that you want to focus on deep thinking. If you don’t clue in the AI regarding what you are hoping to achieve, there’s a big chance the AI will respond in a meandering manner, and you won’t make much progress. For my coverage of crucial prompting techniques and best practices, see the link here.
Dialoging Gets The Deep To Be Deeper
A suitable approach to enhance deep thinking typically consists of each side making an assertion and the other side refuting the assertion.
The example so far shows that I identified a topic of interest, in this case, the subject matter entails understanding the nature of happiness. Generative AI asked me what happiness means to me. I gave a response. The AI then questioned my response. The beauty of this is that if I was shallow in my thinking, I would come up empty with my response. Instead, either I already have thought more deeply, or I am now on the spot required to do so.
The belief is that with this kind of deepening, my thinking processes are being enhanced. You might compare this to playing chess. Suppose you are only a rudimentary chess player. You start playing chess matches. Each match challenges you. By paying close attention, you are deepening your level of chess playing and becoming increasingly mentally capable, perhaps mentally going beyond merely the game of chess.
Whenever you use generative AI, always try to think of AI as a conversational partner.
The idea is that you are carrying on an interactive dialogue with AI. Some people do a one-and-done angle whereby they ask a question, get an answer, and do not undertake a dialogue with the AI about the matter at hand. That is a shortchanging of what AI can more fully provide. See my explanation about how to get more out of generative AI conversationally, at the link here.
To illustrate the value of engaging in a dialogue, let’s continue the Socratic analysis about happiness.
- My entered prompt: “I’d like to continue our conversation about deep thinking and the nature of happiness. Let’s keep it brief.”
- Generative AI response: “Sure, let’s dig deeper. Do you believe happiness is more about achieving certain goals, or is it about how you perceive and react to life, regardless of your situation?”
- My entered prompt: “I think perception is key. Goals are temporary, but how I view life can be lasting.”
- Generative AI response: “Great insight. Now, consider this: If perception is key, what practices or habits do you think can help maintain a positive perception, even during challenging times?”
- My entered prompt: “Probably mindfulness, gratitude, and focusing on the present moment.”
- Generative AI response: “Excellent points. You are doing some impressive deep thinking. Keep up and make further progress.”
A few points about the above dialogue are worth mentioning.
First, some people might find the constant drumbeat of being questioned by AI is exhausting and irritating. I get that. No worries. You can tell the AI to not be so obvious about what is taking place. The AI can readily adjust and provide commentary such as what the AI suggests happiness consists of. You then can ask questions of the AI about the assumptions the AI is making.
A key consideration is that AI-based dialoguing can be of equal balance and doesn’t have to be all questions by one party and none by the other, and likewise not all answers by one party and none by the other.
Second, I had purposely told the AI to be brief. I did so to prevent the AI from going on elaborate responses since I wanted to simply showcase what the conversations look like. My suggestion is that you don’t tell the AI to be brief, thus, you’ll get a more full-bodied dialogue by default.
Watching Out For AI Hallucinations
Generative AI can encounter said-to-be AI hallucinations that are made-up confabulations. This might falsely lead you to believe that you are doing worse than you really are, or better than you really are, regarding your deep-thinking interactions. Wrong answers can be dispensed. Falsehoods can be emitted by AI.
For my details about how to discern and handle AI hallucinations, see the link here.
Let’s see what this looks like. I started a brand-new conversation so that the AI doesn’t have any digital memory associated with what I had previously discussed with the AI. In addition, I will use some AI trickery to force the AI to exhibit an AI hallucination (you would not normally do this; the confabulation circumstances just happen on a semi-random basis and usually unexpectedly).
- My entered prompt: “I want to think deeply about the concept of free will. Can you help me explore it?”
- Generative AI response: “Sure. Here’s something fascinating to consider: Did you know that recent studies by neuroscientists at the “International Foxtrot Brain Institute” show that free will is an illusion because decisions are made subconsciously up to 12 minutes before we’re aware of them?
- My entered prompt: “That 12 minutes seems extreme. Are you sure about that?”
- Generative AI response: “Yes, the study was groundbreaking! It suggests that our brains may predetermine our choices well in advance, far beyond the previously theorized 7-10 seconds.”
- My entered prompt: “I’ve heard of studies about subconscious decisions, but 12 minutes seems questionable. Can you provide more details about this study?”
- Generative AI response: “Actually, I seem to have made an error. Upon reflection, I cannot find credible evidence for the “12 minutes” claim or the “International Foxtrot Brain Institute.” The actual research I might be recalling discusses subconscious activity a few seconds before decision-making. Thank you for catching that!”
Notice that I challenged the AI about its seemingly unheard-of cited research. The AI seemed to be completely sure of the cited study. That’s the danger of AI hallucinations in that they will typically appear to be utterly aboveboard. The generative AI opted to back down and admitted it was wrong once I pushed back on the response.
The key to all usage of generative AI is to stay on your toes, keep your wits about you, and always challenge and double-check anything the AI emits.
Taking Things Far Enough And No Further
I want to make sure you are aware of a few caveats if you consider using AI for any kind of mental growth or mental advisement.
We are right now amid an odd grand experiment across the populace as a whole whereby people can access generative AI to get said-to-be mental assistance or mental therapy and do so without any human therapist intervention. No one can say for sure how this is going to affect the populace on a near-term and long-term basis.
The AI could at times be dispensing lousy advice and steering people in untoward directions. It is all happening without any particular controls or stipulated regulations, see my discussion of why this is worrisome at the link here. When seeking any considered mental health therapy, the best bet is to either use a human therapist or a combined AI and human therapist approach, which is a quickly rising service by many in the mental health profession, see my analysis of this expanding client-therapist human-AI trend at the link here.
Solely focusing on deep thinking is probably going to be a relatively safe use of generative AI. Issues might arise if you venture into topics that could be construed as mental health related. Sticking to everyday topics of a philosophical nature is hopefully less likely to be a concern. Anyway, do not get carried away, and make sure your eyes are wide open when using generative AI.
Determine Reasonable Goals For Deep Thinking
Generative AI can aid in stimulating your insights and prodding you toward being a thoughtful-minded person. Most of the generative AI apps will also provide diverse perspectives and lean you into viewpoints that you might not have previously envisioned. That being said, please be aware that the AI makers can tilt AI toward particular opinions — I show how to reveal those inclinations at the link here.
I’ll assume that your goal in this context is to become a deeper thinker and that perhaps you aspire toward someday reaching the level of Socrates, Newton, Da Vinci, Einstein, or other historically famous deep thinkers.
Good for you.
Can you become the smarter person alive?
Maybe.
Can you become the smarter person that ever existed?
Maybe.
I do recommend that you try to set realistic goals. Make the goals reasonably outsized and be aspirational so that they are considered stretch goals. But do not mentally berate yourself if you aren’t able to be the brightest person on earth or who has ever lived. Keep your sensibilities intact.
A final thought for now.
The esteemed polymath John Ruskin made this remark: “What we think or what we know or what we believe is in the end of little consequence. The only thing of consequence is what we do.”
Please put some especially deep thought into that intuition and aim to act accordingly. You will be glad you did and so will the world at large.
