Morality and psychological development
Spiritual bypass
I’m deeply grateful to everyone that read my last essay and provided thoughtful feedback! Thank you :) You put the wind beneath my sails.
A bunch of people spotted an important gap in the last essay. Specifically, it didn’t actually show the protagonist Rohan grappling with the complex questions of morality surrounding AI. Rohan seemed to make it entirely about him. He found a resolution that seemingly vindicated him no matter what he chose. This is a fair criticism of the essay.
The essay’s original goal was to shed light on the psychological dynamics of why he felt trapped between All In and Staying Put, rather than his specific moral deliberation. To be honest, I also sort of lost my steam partway through writing it. So I ended up leaving the resolution clipped.
Consequently, some were left wondering about the relationship between psychological development and morality. Note that this essay won’t actually show Rohan grappling with his various moral questions directly. That’s quite involved and is well beyond my current energy levels due to some health stuff. But it will slow down enough to conceptually clarify the relationship between morality and psychological development.
Spiritual bypass
John Welwood coined the term spiritual bypass to describe a “tendency to use spiritual ideas and practices to sidestep or avoid facing unresolved emotional issues, psychological wounds, and unfinished developmental tasks.” There are also analogues of developmental bypass, or therapeutic bypass, where one uses therapy-speak to engage in similar patterns.
Almost everyone I know of, including myself, that cares about self-growth has participated in this pattern at least once. Dear reader, I’ve certainly participated in it more than once :)
Although I didn’t design Rohan’s character to bypass, there’s a valid reading where he appeared to do so. Where he continues making everything about him, even at the end. In fact, I wanted him to start as engaging in bypass, and end with becoming more grounded.
At least internally, my test for whether Rohan had actually developed psychologically was that he both felt greater spaciousness, and that he could more skillfully account for the consequences of his actions upon others. But it’s true that the narrative doesn’t conclusively show this side of him!
I hope to rectify some of that with this essay.
How Proving Worth constricted his moral calculus
I’d like to make a distinction between the phenomena that we “see”, and the “frame” that we see them through. For example, the lenses in my glasses allow me to see the world, but they inadvertently mediate and frame what I can see. Rohan’s unconscious stance of Proving Worth was similarly shaping and constraining the possibilities he saw in the world. Specifically, it was framing his engagement with AI as a litigation of his worth.
Psychological development wouldn’t and couldn’t guarantee him becoming a more moral person. But it gave him a better pair of glasses (i.e. Providing Value) with which to see the world. This in turn expanded what felt existentially viable for him. It allowed him to “improve” his moral calculus.
Remember, he wasn’t objectively stuck. But his framing only permitted the mutually exclusive choices of going All In or Staying Put.
That is, as I argued in my previous essay, his internal psychological dynamics constrained which forms of external participation felt existentially viable to him.
Rohan’s underlying arrogance
Rohan had received substantial feedback and validation during his life that he was “smart”. Perhaps smarter than most other people. He’d correctly internalized that these smarts gave him more power than others within specific contexts.
As my coach Brian Whetten often says to me, every strength is paired with a proportionate challenge. In Rohan’s case, his asymmetric power within AI led him to believe that he was, or ever could be, responsible for the whole world. He was correct that “AGI”, whatever that term actually means, would summon immense power within the world. He was incorrect that he was responsible for how it all turned out. Or could even control its trajectory. Especially once the genie was out of the bottle.
This incorrect belief, this arrogance, was underneath his Proving Worth. It was the specific locus of his original spiritual and developmental bypass. The moment he saw he couldn’t actually take responsibility for everything in his own fashion, he was tempted to stay put and take responsibility for nothing.
I’ve consistently found that the most powerful lies we tell ourselves, and therefore others, are those that are true at one level of development, yet false at the next.
Rohan was correct that few people understood frontier AI at a technical level as well as him. Or that his connection to Mario and the rest of the frontier community gave him unique and invaluable perspectives. Perspectives that the rest of the world didn’t yet have. However, he didn’t have a monopoly on insight over the human condition, or how AI should get integrated with it.
His correct estimation of his greater responsibility, proportionate with his greater power, got distorted by his frame (i.e. Proving Worth), into a total responsibility for the world. And he did this in a way that he wasn’t willing to consciously admit to himself or others. No human being, no matter how smart, wise, etc. is built to hold such total responsibility over the world. This inevitably led to his overwhelm, stress and the stuckness that we saw in the essay. It left him torn between the two bad choices of going All In or Staying Put.
Shifting to Providing Value
The shift to Providing Value invited humility over what he could actually hold with integrity. Note that humility here didn’t mean denying his unusual power. That’d just be another form of conceit and bypass.
Humility invited him to take responsibility for a specific piece within the AI ecosystem, rather than attempting to hold for everything by himself. To internalize that his choices mattered, but that they weren’t the only choices that mattered. To build trustworthy relationships with himself and others. Responsibly integrating AI with society would be a team effort between individuals, companies and countries, of which he would only contribute a relatively small piece. No matter how smart, accomplished or wise he was or would ever become.
Mario’s invitation to Rohan was but one such relational invitation available to him. He could have sought out and tested other such relationships via the Enrollment process described in the previous essay. He could have even tested entering into honest relationships with his old coworkers within Hooli’s Apollo program.
This is where the moral calculus came in, which I entirely omitted from the previous essay.
A deeper stance of humility via Providing Value didn’t and couldn’t automatically tell him what was morally acceptable. But it did create enough spaciousness to make existentially viable a fuller spectrum of action between All In and Staying Put. He could start his investigation with high-level purposes he felt were good for the world, viable for Noetic, but crucially tempered with what he could make himself specifically responsible for.
This involved using both his brain and his heart to read extensively. To do his research and understand the space of plausible actions and consequences. It’s true that moral certainty is impossible for a human being limited in space and time. But that didn’t release him from the obligation both to himself and others in carefully justifying his choices. He’d eventually need to take an intuitively aligned and well-reasoned leap of faith. Faith in himself, his relationships with others and the world at large. But the emphasis is on intuitively aligned and well-reasoned. Faith without structures of self-correction would have been yet another form of bypass.
I skipped all this deliberation and soul-searching in the essay and jumped to his choice of “red-teaming”. But it’s an absolutely critical part of the process.
As mentioned in the previous essay, moral, technical and developmental framings are recursively related. Moral inquiry helped him judge alignment with the Good. Technical inquiry allowed him to judge a choice’s feasibility. And psychological development expanded the space of possibility that was existentially viable to him.
Growing the capacity to see greater possibility isn’t always the same as becoming better calibrated on the “goodness” of each choice.
Rohan lacked sufficient capacity for “leadership”
As Brian often tells me, leadership is the capacity to translate power into specific, accountable value. To engage in such translation honestly about the inherent certainties, uncertainties and externalities with wielding power. Rohan was stuck partly because he lacked sufficient leadership skills to translate power into specific value. The process of cultivating such leadership, or the difference between “better” and “worse” forms of leadership is well outside the scope of this essay.
Consciously testing one’s provisional moral calculus is also an act of leadership that Rohan originally lacked. For example, he got too easily got swept away by the galaxy-brained thinking of others in the AI community. Despite having seen every season of Silicon Valley, he easily got excited about using AI to “make the world a better place” in a bypass-ey kind of way.
The humility within Providing Value invited Rohan to ask Mario what specific value they’d be able to create together, and the externalities they both foresaw. It invited Rohan to consciously explore how Mario/Rohan could build trust and commitment with each other, both in terms of their interpersonal relationship, and Noetic’s actions within the world. I skipped all this when I talked about Enrollment. I vaguely gestured at it when I talked about Win/Win or No Deal relationships.
Rohan’s core question
Ultimately, psychological development invited Rohan to shift from asking himself, “Am I good or worthy enough?”, to asking himself something like “Given my specific powers and abilities in the world, and the invitations in front of me, what specific outcomes am I willing to make myself answerable for with integrity?”
The former question was largely about himself. The latter provided him with the spaciousness to examine the specific moral and technical ramifications of his participation in the AI ecosystem, as a living and self-corrective relationship with the world.