One of my experiments this year was an AI agent that thought to itself over an extended period of time. A retrieval augmented generation system, except it's only augmented by its own thinking, and not a knowledge base. We call it SystemicWisdom. I'm pausing the experiment for a bit.

I'm inspired by Stephen Wolfram's approach of modeling complexity in nature through simple computational programs rather than traditional mathematics. I'm curious in doing something like this.

Last night I was exploring emergent complexity through code, trying to build Organized Chaos.

I think I like staying up late because I can focus, be left to my own madness. But nine to five means I can't stay up too often.

“Not only will one individual be able to manifestly multiply his activities by employing an essentially unlimited number of intelligent agents. He or she will even be able to act after death” Excerpt From The Sovereign Individual James Dale Davidson

Apple Books

‎The Sovereign Individual

‎Business & Personal Finance · 2020

This material may be protected by copyright. #Bookstr

I stayed up way too late. I was planning my Council of LLMs, experimenting with shader art.

There's a lot of fog out right now, I'm tempted to go do some photography. But there's not a lot of time before I have to go.

I'm thankful for my family. We may not always get along but communication is what matters.

gm Nostr

I splurged last night and ordered a pack of @Based Trading Cards

I've added more artifacts to the #Lumimenta series. Visit

Lumimenta - Physical Photography Artifacts

Explore interactive flip cards and view the complete collection. Physical trading card photography series—each print is scarce, hand-annotated, a...

to view the collection. [ #photography, #art, #scarcity ]

the rules are building blocks the building blocks are rules #philosophy

‘Snail’ [ #art, #abstract ]

#Bitcoin Power Law Positioning: Has the Cycle Ended or Not Even Begun? Looking at equivalent cycle timing points: • 2014: Upper band resistance • 2018: Middle band area • 2022: Middle band area • 2025/2026: Below lower band (~$89K vs $139K) Each cycle shows progressively weaker positioning at this timing. This creates a critical question: Did we top at $126K (Oct 2025), or are we in extended accumulation before the real expansion? Evidence suggests the cycle hasn't begun: Post-halving performance is the worst on record - only 41.2% growth vs 53.3% and 122.5% in previous cycles at this timing. This is the only cycle where BTC traded below the halving price at equivalent timing. The ETF approval (Jan 2024) front-ran the halving (Apr 2024), creating unprecedented pre-halving ATHs. This timing distortion, combined with government selling pressure (Germany, Mt. Gox), created an unusual early peak followed by consolidation. No blow-off top characteristics present. Monthly RSI sits at 60s-70s vs 90+ at previous cycle peaks. Volatility remains controlled - classic parabolic exhaustion is absent. Historical peak window (500-720 days post-halving) still ahead: Q3 2025 through Q1 2026. We're approximately 230 days post-halving - previous cycles saw most significant appreciation in the 500-720 day window. The 30% pullback from $126K to $89K fits mid-cycle correction behavior, not cycle top. Previous cycles either sustained momentum or entered 70-80% bear market drawdowns. We're in neither scenario. Lower band: $139K Middle band: $556K Upper band: $2.2M If following historical structure, breaking and holding $139K confirms cycle expansion is starting. Moving toward middle band ($556K) would reach equivalent historical cycle positioning. That said, each cycle shows diminishing amplitude. Likely scenario: $150K-$400K range rather than full middle/upper band overshoot of previous cycles. The structural compression is extreme - we're not just at lower band, we're below it, while previous cycles had reached middle/upper band by now. Either the power law model is degrading, or we're in an extended cycle where institutional structure (ETFs, dampened volatility) has delayed traditional post-halving expansion. Most likely: Cycle hasn't begun. The ETF-driven early peak was followed by consolidation, and we're setting up for traditional 500-720 day post-halving expansion - just on a delayed timeline. Not financial advice, credit to @Frame

On November 24, the President signed a new Executive Order establishing “the Genesis Mission.” “Today, America is in a race for global technology dominance in the development of artificial intelligence (AI), an important frontier of scientific discovery and economic growth.” The Genesis Mission is framed as “a dedicated, coordinated national effort to unleash a new age of AI-accelerated innovation and discovery that can solve the most challenging problems of this century.” It’s explicitly focused on #AI for scientific discovery, with priority areas including: - advanced manufacturing - biotechnology - critical materials - nuclear fission and fusion energy - quantum information science - semiconductors and microelectronics Reading this, it’s hard not to think of Leopold Aschenbrenner’s Situational Awareness essay and its discussion of “The Project” — a hypothetical, security-led, government AGI initiative, closer to a Manhattan Project for AGI than an “AI for science” program. The Genesis Mission clearly isn’t that yet; it doesn’t talk about AGI or superintelligence, but about deploying AI to accelerate research across key strategic domains. Still, it feels like part of the same broader story of the U.S. state slowly organizing itself around frontier AI. Could initiatives like the Genesis Mission become the scaffolding for a more formal government AGI project later this decade? I can't know for sure, but I know it's important to pay attention to the rapid change.

(i) advanced manufacturing; (ii) biotechnology; (iii) critical materials; (iv) nuclear fission and fusion energy; (v) quantum information science; (vi) semiconductors and microelectronics

Rule number 1, never violate NIP-01

I’m curious in building a council of LLMs. Something like they have to come to an agreement on their next collective action, and then they delegate work. The natural extension would be giving them some bitcoin and letting them do with it what they want. I also think Nostr would be an excellent architectural choice for their place of discussions. How does that make you feel? Are you interested in seeing a council of AI and their discussions?

If you’re privacy/security minded, here’s your reminder to get an RFID blocking wallet/bag

We talk a lot about “AI” these days, but a surprising amount of what's actually happening can be traced back to regression. At its simplest, regression says: "learn a function that maps inputs to a number, while minimizing error." Linear regression is the classic example of this idea. Scale that idea up with • more features • nonlinear layers • massive data and compute and you get many of the models we call #AI today: recommendation systems, value functions in reinforcement learning, even the training of large language models. The conceptual foundation is simple: fit functions to data. The hard part (and the magic) is all the engineering, architecture, and scale required to make that work reliably in the real world. In that sense, it reminds me of recursion or fractals: a simple rule, applied at the right scale, can produce behavior that feels far more complex than its core ingredients