On January 4, 2021, Sam Altman, the CEO of OpenAI, the leading Artificial General Intelligence (AGI) company in the world, tweeted the following question… If an oracle told you that human-level AGI was coming in 10 years, what about your life would you do differently?
For now it seems, we push the buttons for AI. Soon it will start pushing its own. I believe all hell breaks loose when it learns to push our buttons. Imagine you're a great writer (which I'm not) and how demoralizing it will be if/when AI is far better than you.
You expect a computer to play chess better, but I think it's going to be something much different when it writes a poem that makes you cry.
Hey Mark - Agreed! One of Altman's predictions is...
"i expect ai to be capable of superhuman persuasion well before it is superhuman at general intelligence, which may lead to some very strange outcomes"
This article was mind-blowing. I'm letting the impact of what you wrote about percolate in mind and later when I get the chance I'll write you a really good comment! But for now just know that you set off shock waves in my brain.
Love the idea of a Manhattan Project for human potential! Are you familiar with the writings of Sri Aurobindo and his integral vision for the future of humanity? I see Sri Aurobindo and his vision as absolutely crucial now to navigate the current paradigm and awaken the superhuman within us.
Can we see every word as coming from a meta intelligence that is trying to self organize at every single moment? There is a screen between us but there is also an infinite compassionate force holding the conversation and rewiring our collective synapses with every interaction. Maybe facing this threat the impulse to self organize at the meta level is drastically quickening. Can you feel it? A type of light speed intelligence that is operating in the quantum field, and somewhere deep in the heart, that is organically seeking a solution to the AI conundrum.
In either case I deeply agree that the threat can be a massive catalyst to extract our deepest potential and the hidden power of the human heart and Soul.
Michael, thank you putting this out there. I too go back and forth on how to react to/think about what is coming. I don't have a clear answer to "how best to prepare."
You were very candid in your article about this. But based on your insight in this topic, I'd be curious to learn precisely what you're thinking in term of the "how" of Human Superintelligence. I get what it could be applied to based on your list. But are you envisioning and thinking of prioritizing direct interface (e.g. Neuralink), or are you more focused on simply leveraging tools to upgrade our collective knowledge and consciousness?
At the moment, I'm thinking of, "How can we use AI NOW to upgrade our collective knowledge and consciousness?" The keywordi is NOW. This would start with using AI tools with the explicit purpose of augmenting human intelligence rather than than just automating it?
That said, I'm still in the very early stages of thinking on this. It's just a pathway that I'm excited to explore and be creative/imaginative with.
Deeply appreciative of this gathering of information and thoughts. Such a massively important topic. I am left wondering two things. What are the biggest life extinguishing threats predicted AND what does rethinking your life from first principles look like? (Your adaptation 4). Presently at cafe so couldn’t watch videos. Will do when I get home. Again. Thankyou
Once you watch the videos (even just the short ones) you'll be able to start imagining the scale of the threat. I think pretty much nothing is off the table of possibilities. The thing is that not only will it be 1000x smarter, it will be 1000x faster. It'll checkmate us in the blink of an eye. I can literally visualize concentration camps where humans do physical labor in order to sustain and be sustained by the AI because it can out-compete and out-think us at every step. Even now we are already dependent on AI as you will see in the Nvidia video. We couldn't just pull the plug to save ourselves without creating a new Dark Age.
I see the key as using ever-improving AI to help solve the problem of AI safety. Use its own smarts to stop it from becoming evil.
Wow, this article just changed so much for me. Thank you very much for such a clear and imaginable explanation. I will be researching about this topic a lot more now.
I don't know of any specific resources about different cultures and AI, only that, through deep reading about the history of the state in China, I'd predict that its better equipped than the West in dealing with such complex, values-based issues when they impact the broader society.
As a Westerner sitting in China and witnessing all the fear in the West, I can't help but laugh at how little the West understands China, even after hundreds of years., so when you talk about imagining the future of AI and its impacts on humanity, it seems hopeless. It was only in the last 20 years that psychologists have seriously considered China, as a civilization, might think differently, and discovered that it does and always has. An example is surgery. While surgery has been practiced in many different Western societies for thousands of years, it has been very rare in China despite China's advances in so many other areas, as a result of the different kind of thinking it typically has. The difference is explained by the more wholistic thought processes of Chinese who have a naturally difficult time with the idea of isolating a single body part from the body within which it exists and the myriad body parts to which it is intimately connected. Westerners, in contrast, have little problem.
I think something similar can be said about AI and its impacts on society/humanity overall and think the typical holistic thought processes of Chinese are better adapted to coping with such big questions (especially when it involves change) compared to our own typically narrow and more rigid, Western thinking, so my argument would be to make sure this kind of thinking is included. In short, we're all here as individualistic thinkers, to figure out how we can benefit from AI whereas, the Chinese collective approach would first focus on how to ensure benefits to society.
In conclusion, as with all technology, AI is a tool and isn't inherently good or evil, that's determined by the moral leadership of a society that implements it. In other words, its a values, not a technical question. That's why its hard not to believe that Chinese civilization, with its 2000 years of hiring civil servants based on morals and ethics and merit is in a superior position when dealing with such questions..
Dec 26, 2023·edited Dec 26, 2023Liked by Michael Simmons 🪵🔥 ⛺
Hi Michael. Thanks for this article. It brings together a lot of the aspects about AI and its current developments in a way that’s understandable for a broad audience. And the more people thinking about this, the better.
I thoroughly enjoyed this read. I’m not going to attempt to speak to every point, but two things stick out to me:
1. I enjoy learning and discovering, and I enjoy learning from other humans who are learning and discovering. AI may be able to deliver the answers faster than another human can... but real understanding on our part comes through learning, pondering, debating, considering, evaluating, and on and on... I think we will always crave that part of the learning/discovery process, and I don’t think AI will ever be able to comprehend that in a way that it can deliver it to us as effectively as a human teacher can.
2. No matter how smart it gets - and even if it someday can replicate emotion - AI will never “get” what it means to be human, because it will never “live” with the fear and the certainty of its own impending death. It will never comprehend what it’s like to be mortal. That will protect those workers/creators who show us our mortality, and the patrons who want to be shown.
Everyone else might be doomed to a life of ease and convenience though; I think your guess is as good as anyone’s when it comes to that.
Michael Glenn, it seems that another thing that AI won't be able to do, related to both your excellent points, is create and lead a team or a community. The human vulnerabilities, mistakes, messy learning processes, fear of failure, and joys of collaboration and incremental success are all part of that process.
That said, it's plain to see that AI agents will be complementary players for teams and communities.
Love the HSI idea. Partly in response to AI, I’ve started spending a lot more time developing athletic skills. I reason that there are body-based reasoning skills (motor skills + metaphors) that AI like GPT don’t have much access to. Furthermore, I wonder what ways robotics could be used for enhancing athletic training (and thus reasoning ability). Personally, I find myself to be more practical and intuitive in my intelligence now, due to my recent athletic training. I’m basically training my intuition on how reality works, in painstaking, sensory-rich detail.
David K, focusing on body-based reasoning skills in the age of AI is an insight I'd not considered as a mountain biking skills coach. So thank you.
I asked Anthropic's Claude.ai this question, from a recent Forbes article titled "5 ChatGPT Prompts To Become A Sought-After Thought Leader."
"My current role is a mountain biking skills coach who specializes in helping my clients learn deliberate practice techniques. I'm looking to expand my skill set in ways that set me apart from the competition in the mountain biking skills coaching industry. What are the emerging skills or knowledge areas in my industry that I should focus on learning to maintain a cutting-edge position and maximum relevance with my clients? The aim is to identify areas where I can gain new expertise that not only differentiate me but also allow me to share unique insights and teachings with others, thereby reinforcing my position as a thought leader."
Claude's reply was helpful to me both as a mountain biker trying to improve and a coach.
Let me know what you think. (And you, too, Michael Simmons, since thought leadership is right up your alley!) Claude wrote:
============
"Here are some recommendations for emerging skills and knowledge areas to help differentiate you as a mountain biking deliberate practice coach:
1. Neuroscience of Motor Learning - Study latest research on how the brain acquires and retains motor skills. Apply to biking mastery.
2. Performance Analytics - Leverage wearable sensors, video analysis tools to quantify skill levels and micro-changes undetectable to naked eye. Allows precision tracking of subtle progress.
3. Virtual Reality Applications - With VR tech advancing, explore using simulated environments for skills training without real-world consequences. Creates hyper-focused practice.
4. Adaptive/Personalized Instruction - Use AI and machine learning to analyze clients' strengths/weaknesses and customize training plans, like a virtual coach adapting in real-time.
5. Mental Fitness - Expand your knowledge of high performance psychology. Mindfulness, stress management, cognitive behavioral approaches to build skills like resilience, focus, optimal arousal levels.
6. Contemplative Practices - Study how activities like journaling, meditation, even time in nature boost deliberate practice capabilities by supporting metacognition and introspection.
7. Skill Transfer Theory - Research the science around transferring deliberate practice proficiency between domains. Mountain biking shares neuromechanics with many sports for cross-application.
Focusing on these emerging areas allows you to translate cutting edge knowledge into distinct coaching advantages, cementing your position as a forward-thinking leader in applying deliberate practice to mountain biking mastery. Let me know if you need any other recommendations!"
Nice tips -- I like 1 and 2 especially. I’m imagining using wearables to collect data and send signals to an athlete to adjust their movement or posture in real-time, in order to build the skills and intuition more rapidly.
I watched this video two years ago, and it changed the way I think about education. The video touches upon many points you mentioned in your article. You might enjoy the video and the paper that came with it. The link for the paper no longer works, but I'm happy to share it if you're interested.
Thank you Michael for another thorough and thought provoking article!
I appreciate your breakdown of AI response options.This prompted my own reflections, and though I feel over my head on the topic, the urgency resonates so I’ll chime in.
I loved your point on the common, fixed mindsets about love, compassion, wisdom, etc., and the crucial opportunity for further development. It’s also true that many spiritual traditions, especially Eastern, have for centuries found and replicated experiences of human love/wisdom that are almost incomprehensible to the average human. And, they’ve offered detailed guidance for how anyone obtain this knowledge.
It’s not clear to me how AI would help humans to access this wisdom, except to the extent that the urgency of AI's existential threat is taken seriously.
More likely, it seems to me that AI could be a huge distraction from the change in consciousness that humanity needs.
What material science has recently discovered about the value of meditation (e.g. it reduces stress), for example, is infinitesimal compared to the vastness of Buddhist teachings on “emptiness” and “no-self.”
It’s one thing to identify and study measurable, physical correlates to inner states of being that sages have described, but it’s another to conclude that such states can be reduced to measurements. Is there evidence that AI, no matter how fast, could be capable of anything beyond the mere appearance of compassion?
My impression is that AI is being designed and hyped mostly by people who DO have fixed mindsets about intelligence—at least to the extent that intelligence is being defined in terms of speed and measurable computational ability. But is that definition accurate?
In Pali (the language spoken by the historical Buddha) there is one word for heart-mind, “Citta,” and another word for intellect: Manas.
The world-renown spiritual teacher Jiddhu Krishnamurti said: “there is no intelligence without love…And love is not the movement of thought.”
The idea of AI progress I’ve heard a lot of seems to require being convinced, as Neil Postman put it, that “technical progress is humanity’s supreme achievement and the instrument by which our most profound dilemmas are solved.”
It’s not clear to me that this is true, even if I concede that technology can have enormous benefits.
I so appreciate that you’re widening the lens on this discussion.
For now it seems, we push the buttons for AI. Soon it will start pushing its own. I believe all hell breaks loose when it learns to push our buttons. Imagine you're a great writer (which I'm not) and how demoralizing it will be if/when AI is far better than you.
You expect a computer to play chess better, but I think it's going to be something much different when it writes a poem that makes you cry.
Hey Mark - Agreed! One of Altman's predictions is...
"i expect ai to be capable of superhuman persuasion well before it is superhuman at general intelligence, which may lead to some very strange outcomes"
https://twitter.com/sama/status/1716972815960961174
This article was mind-blowing. I'm letting the impact of what you wrote about percolate in mind and later when I get the chance I'll write you a really good comment! But for now just know that you set off shock waves in my brain.
Yes, it is a lot to hold in one's mind!!!
Love the idea of a Manhattan Project for human potential! Are you familiar with the writings of Sri Aurobindo and his integral vision for the future of humanity? I see Sri Aurobindo and his vision as absolutely crucial now to navigate the current paradigm and awaken the superhuman within us.
Can we see every word as coming from a meta intelligence that is trying to self organize at every single moment? There is a screen between us but there is also an infinite compassionate force holding the conversation and rewiring our collective synapses with every interaction. Maybe facing this threat the impulse to self organize at the meta level is drastically quickening. Can you feel it? A type of light speed intelligence that is operating in the quantum field, and somewhere deep in the heart, that is organically seeking a solution to the AI conundrum.
In either case I deeply agree that the threat can be a massive catalyst to extract our deepest potential and the hidden power of the human heart and Soul.
John - I have his book, but I haven't read it. But your message is a great impetus to do so. So thanks!
I love the way you stated things. Just subscribed to you. I can feel that possibility of collective awakening in a more than just intellectual way.
Michael, thank you putting this out there. I too go back and forth on how to react to/think about what is coming. I don't have a clear answer to "how best to prepare."
You were very candid in your article about this. But based on your insight in this topic, I'd be curious to learn precisely what you're thinking in term of the "how" of Human Superintelligence. I get what it could be applied to based on your list. But are you envisioning and thinking of prioritizing direct interface (e.g. Neuralink), or are you more focused on simply leveraging tools to upgrade our collective knowledge and consciousness?
Good questions.
At the moment, I'm thinking of, "How can we use AI NOW to upgrade our collective knowledge and consciousness?" The keywordi is NOW. This would start with using AI tools with the explicit purpose of augmenting human intelligence rather than than just automating it?
That said, I'm still in the very early stages of thinking on this. It's just a pathway that I'm excited to explore and be creative/imaginative with.
Makes sense. Thanks!
Deeply appreciative of this gathering of information and thoughts. Such a massively important topic. I am left wondering two things. What are the biggest life extinguishing threats predicted AND what does rethinking your life from first principles look like? (Your adaptation 4). Presently at cafe so couldn’t watch videos. Will do when I get home. Again. Thankyou
Once you watch the videos (even just the short ones) you'll be able to start imagining the scale of the threat. I think pretty much nothing is off the table of possibilities. The thing is that not only will it be 1000x smarter, it will be 1000x faster. It'll checkmate us in the blink of an eye. I can literally visualize concentration camps where humans do physical labor in order to sustain and be sustained by the AI because it can out-compete and out-think us at every step. Even now we are already dependent on AI as you will see in the Nvidia video. We couldn't just pull the plug to save ourselves without creating a new Dark Age.
I see the key as using ever-improving AI to help solve the problem of AI safety. Use its own smarts to stop it from becoming evil.
Wow, this article just changed so much for me. Thank you very much for such a clear and imaginable explanation. I will be researching about this topic a lot more now.
Out of curiosity, what changed for you?
I mean I take this topic as far more urgent and serious than I thought it is
I don't know of any specific resources about different cultures and AI, only that, through deep reading about the history of the state in China, I'd predict that its better equipped than the West in dealing with such complex, values-based issues when they impact the broader society.
As a Westerner sitting in China and witnessing all the fear in the West, I can't help but laugh at how little the West understands China, even after hundreds of years., so when you talk about imagining the future of AI and its impacts on humanity, it seems hopeless. It was only in the last 20 years that psychologists have seriously considered China, as a civilization, might think differently, and discovered that it does and always has. An example is surgery. While surgery has been practiced in many different Western societies for thousands of years, it has been very rare in China despite China's advances in so many other areas, as a result of the different kind of thinking it typically has. The difference is explained by the more wholistic thought processes of Chinese who have a naturally difficult time with the idea of isolating a single body part from the body within which it exists and the myriad body parts to which it is intimately connected. Westerners, in contrast, have little problem.
I think something similar can be said about AI and its impacts on society/humanity overall and think the typical holistic thought processes of Chinese are better adapted to coping with such big questions (especially when it involves change) compared to our own typically narrow and more rigid, Western thinking, so my argument would be to make sure this kind of thinking is included. In short, we're all here as individualistic thinkers, to figure out how we can benefit from AI whereas, the Chinese collective approach would first focus on how to ensure benefits to society.
In conclusion, as with all technology, AI is a tool and isn't inherently good or evil, that's determined by the moral leadership of a society that implements it. In other words, its a values, not a technical question. That's why its hard not to believe that Chinese civilization, with its 2000 years of hiring civil servants based on morals and ethics and merit is in a superior position when dealing with such questions..
Interesting. Thanks for sharing Mark.
I haven't gone deep on reflecting about how different cultures are thinking about AI. Are there any resources on this this you recommend checking out?
Hi Michael. Thanks for this article. It brings together a lot of the aspects about AI and its current developments in a way that’s understandable for a broad audience. And the more people thinking about this, the better.
Since you spend a lot of time and energy on your research, I was wondering if you already came across Daniel Schmachtenberger. His main theme is the Meta Crisis (https://youtu.be/4kBoLVvoqVY?si=FKHKbiCCPV2UtQls 50min) and he sees a big role for AI in either accelerating or solving it (https://youtu.be/4kBoLVvoqVY?si=FKHKbiCCPV2UtQls 3:12h).
His Consiliance Project might be a form of (decentralized) Manhattan Project (https://consilienceproject.org/about-the-project/)
Thanks Rein! Yes, familiar with Daniel and a fan of him and his work.
Awesome
I thoroughly enjoyed this read. I’m not going to attempt to speak to every point, but two things stick out to me:
1. I enjoy learning and discovering, and I enjoy learning from other humans who are learning and discovering. AI may be able to deliver the answers faster than another human can... but real understanding on our part comes through learning, pondering, debating, considering, evaluating, and on and on... I think we will always crave that part of the learning/discovery process, and I don’t think AI will ever be able to comprehend that in a way that it can deliver it to us as effectively as a human teacher can.
2. No matter how smart it gets - and even if it someday can replicate emotion - AI will never “get” what it means to be human, because it will never “live” with the fear and the certainty of its own impending death. It will never comprehend what it’s like to be mortal. That will protect those workers/creators who show us our mortality, and the patrons who want to be shown.
Everyone else might be doomed to a life of ease and convenience though; I think your guess is as good as anyone’s when it comes to that.
Michael Glenn, it seems that another thing that AI won't be able to do, related to both your excellent points, is create and lead a team or a community. The human vulnerabilities, mistakes, messy learning processes, fear of failure, and joys of collaboration and incremental success are all part of that process.
That said, it's plain to see that AI agents will be complementary players for teams and communities.
Love the HSI idea. Partly in response to AI, I’ve started spending a lot more time developing athletic skills. I reason that there are body-based reasoning skills (motor skills + metaphors) that AI like GPT don’t have much access to. Furthermore, I wonder what ways robotics could be used for enhancing athletic training (and thus reasoning ability). Personally, I find myself to be more practical and intuitive in my intelligence now, due to my recent athletic training. I’m basically training my intuition on how reality works, in painstaking, sensory-rich detail.
David K, focusing on body-based reasoning skills in the age of AI is an insight I'd not considered as a mountain biking skills coach. So thank you.
I asked Anthropic's Claude.ai this question, from a recent Forbes article titled "5 ChatGPT Prompts To Become A Sought-After Thought Leader."
"My current role is a mountain biking skills coach who specializes in helping my clients learn deliberate practice techniques. I'm looking to expand my skill set in ways that set me apart from the competition in the mountain biking skills coaching industry. What are the emerging skills or knowledge areas in my industry that I should focus on learning to maintain a cutting-edge position and maximum relevance with my clients? The aim is to identify areas where I can gain new expertise that not only differentiate me but also allow me to share unique insights and teachings with others, thereby reinforcing my position as a thought leader."
Claude's reply was helpful to me both as a mountain biker trying to improve and a coach.
Let me know what you think. (And you, too, Michael Simmons, since thought leadership is right up your alley!) Claude wrote:
============
"Here are some recommendations for emerging skills and knowledge areas to help differentiate you as a mountain biking deliberate practice coach:
1. Neuroscience of Motor Learning - Study latest research on how the brain acquires and retains motor skills. Apply to biking mastery.
2. Performance Analytics - Leverage wearable sensors, video analysis tools to quantify skill levels and micro-changes undetectable to naked eye. Allows precision tracking of subtle progress.
3. Virtual Reality Applications - With VR tech advancing, explore using simulated environments for skills training without real-world consequences. Creates hyper-focused practice.
4. Adaptive/Personalized Instruction - Use AI and machine learning to analyze clients' strengths/weaknesses and customize training plans, like a virtual coach adapting in real-time.
5. Mental Fitness - Expand your knowledge of high performance psychology. Mindfulness, stress management, cognitive behavioral approaches to build skills like resilience, focus, optimal arousal levels.
6. Contemplative Practices - Study how activities like journaling, meditation, even time in nature boost deliberate practice capabilities by supporting metacognition and introspection.
7. Skill Transfer Theory - Research the science around transferring deliberate practice proficiency between domains. Mountain biking shares neuromechanics with many sports for cross-application.
Focusing on these emerging areas allows you to translate cutting edge knowledge into distinct coaching advantages, cementing your position as a forward-thinking leader in applying deliberate practice to mountain biking mastery. Let me know if you need any other recommendations!"
Nice tips -- I like 1 and 2 especially. I’m imagining using wearables to collect data and send signals to an athlete to adjust their movement or posture in real-time, in order to build the skills and intuition more rapidly.
Agreed. For example, a wearable device that sent signals to me to pay closer attention to my hip hinge while mountain biking would be quite helpful
https://www.youtube.com/watch?v=DzL4RpHnY44
I watched this video two years ago, and it changed the way I think about education. The video touches upon many points you mentioned in your article. You might enjoy the video and the paper that came with it. The link for the paper no longer works, but I'm happy to share it if you're interested.
To be honest, when I saw your title my heart sank. I knew immediately what it was about and I thought, "Why did we let this genie out of the bottle?"
By the end of your post I was a little depressed. How much sci-fi can we crunch through in one decade? And I'm old. What about the 25 year olds?
It reminds me of that line when Morpheus tells Neo, "it was the end of the 20th century, when mankind marveled at its own creation, the birth of AI."
Ha, ha! Can't wait for the digital knock when whatever intelligence decides, "yeah, we're in charge now. And your bank accounts too!"
An awesome article on some distressing possibilities.
Thank you Michael for another thorough and thought provoking article!
I appreciate your breakdown of AI response options.This prompted my own reflections, and though I feel over my head on the topic, the urgency resonates so I’ll chime in.
I loved your point on the common, fixed mindsets about love, compassion, wisdom, etc., and the crucial opportunity for further development. It’s also true that many spiritual traditions, especially Eastern, have for centuries found and replicated experiences of human love/wisdom that are almost incomprehensible to the average human. And, they’ve offered detailed guidance for how anyone obtain this knowledge.
It’s not clear to me how AI would help humans to access this wisdom, except to the extent that the urgency of AI's existential threat is taken seriously.
More likely, it seems to me that AI could be a huge distraction from the change in consciousness that humanity needs.
What material science has recently discovered about the value of meditation (e.g. it reduces stress), for example, is infinitesimal compared to the vastness of Buddhist teachings on “emptiness” and “no-self.”
It’s one thing to identify and study measurable, physical correlates to inner states of being that sages have described, but it’s another to conclude that such states can be reduced to measurements. Is there evidence that AI, no matter how fast, could be capable of anything beyond the mere appearance of compassion?
My impression is that AI is being designed and hyped mostly by people who DO have fixed mindsets about intelligence—at least to the extent that intelligence is being defined in terms of speed and measurable computational ability. But is that definition accurate?
In Pali (the language spoken by the historical Buddha) there is one word for heart-mind, “Citta,” and another word for intellect: Manas.
The world-renown spiritual teacher Jiddhu Krishnamurti said: “there is no intelligence without love…And love is not the movement of thought.”
The idea of AI progress I’ve heard a lot of seems to require being convinced, as Neil Postman put it, that “technical progress is humanity’s supreme achievement and the instrument by which our most profound dilemmas are solved.”
It’s not clear to me that this is true, even if I concede that technology can have enormous benefits.
I so appreciate that you’re widening the lens on this discussion.
-Chris
I concur completely!