Each of us is a wonderful and bewildering bundle of doing, being, and becoming; predicting, planning, and optimizing; acting and reacting; feeling, hurting, and enjoying; developing and degrading; observing, attending, knowing, and forgetting; thinking, framing, interpreting, rationalizing, and bullshitting; and on and on.

I’m something of an authority on what all of these dynamics equate to within my own being. But not an infallible authority (or always even the best authority), as I must observe, attend, frame, and interpret in order to say anything at all about them.

When I look at myself, here is one of the narratives that emerge:

The world is full of problems. Much of my life has been involved in allowing my allowing myself to be driven by my curiousity about some of these problems. But lately, I’ve begun seeking not just to entertain my curiosity but rather to work on the problem of pursuing a deep and intentional form of integration.

To seek integration is to seek greater alignment and coherence among the parts of the systems that make up each of our worlds—systems like the system of my bodymind and its parts, systems of relationships and social networks, systems of institutions, and very soon, systems of artificial minds, their parts, their relationships, their institutions.

To state something that is obvious (once it is obvious): by default, the human dynamical bundle is very leaky, incoherent, and unintegrated. Which is to say that while there are some feedback processes which naturally tend to drive all the different parts of the bundle into some kind of alignment, all of these mechanisms are incomplete and imperfect; I can care deeply about one thing and yet be doing little within my power to effect that thing.

Within this general theme, here are some of the problems that are currently salient to me:

Problem Framing. I’m interested in how problems are solved through the joint processes of problem framing and optimization within a frame. If we are stuck, stagnating, or ineffective in life, often it is because we are operating from within a defective frame. Seeking “full-stack” integration is itself a frame-shift for many of us who are stuck trying to make ourselves happy by effecting changes only in the world, rather than at a point of higher leverage–within the parts of ourselves. I believe the pathway to integration to be littered with frame-shifts.

AI interpretability / alignment: Human intelligence is facilitated by attention flows, which serve to coordinate the massive distributed capacity of the mind’s parts. These flows occur within a representation space which is interpretable, at least to other parts of the mind. Supposing this pattern of “attention as coordination layer” is essential to higher intelligence, how can it be useful in improving efforts toward the interpretability / alignment of modern AI architectures broadly viewed as inscrutable?

The metacrisis: The best of all tautologies is Darwin’s: To survive, something must be not thoughtful, virtuous, beautiful, good, or anything at all except good at surviving. When the environment is dense with broad spectrum of possibility (a big world in which anything that can be done will be done), this makes it critical that there is good alignment between what it means to be good at surviving and what it means to be good. The lesson of Moloch is that this is often not the case (See Twitter for a microcosm of this). But things which are loving, virtuous, forgiving, abd cooperative often have an advantage for survival. What technologies (including psychotechnologies) are needed to amplify this advantage? Or more generally, what structural changes to society can be enacted to strengthen this alignment?

Personal development: Finding freedom and releasing double-binds by pursuing integrity.