The “thing” I’m showing is the mapping itself: it separates two questions that often get conflated: (1) how structurally aligned a public entity is with long-horizon AI value creation, and (2) how much of that story already appears to be priced in.
The x-axis (M.I.N.D.) is a composite structural-alignment score (Material, Intelligence, Network, Diversification, inspired by the “Last Economy” framing). Scores are synthesized per entity after a skills/assets/capabilities analysis and a review of analyst research, using an LLM as a structured aggregation tool rather than an oracle. Roughly speaking: Material captures control over scarce physical inputs, Intelligence reflects leverage over computation and models, Network captures ecosystem and data flywheels, and Diversification reflects exposure across multiple AI value paths.
The y-axis (valuation tension) is a rough proxy for expectation saturation. I’m treating it as a secondary signal; the primary thing I’m testing is whether structural alignment and narrative intensity decouple in interesting ways.
One weakness I’m actively unsure about is the M.I.N.D. formulation itself. Multiplying the four dimensions strongly penalizes any missing leg, which may or may not reflect how value actually compounds in AI systems. If that assumption is wrong, the framework will systematically mislead.
I’m especially interested in: - whether these four dimensions are the right ones - whether multiplication is the right way to combine them - where this framework would clearly fail
Happy to answer questions or clarify assumptions.
M.I.N.D. / "Last Economy"
The "Last Economy" framing comes from Emad Mostaque's book of the same name and is a way of thinking about where long-run value concentrates when intelligence becomes abundant. M.I.N.D. is the operationalization of that idea from the book and positioned as a better "yardstick" than current metrics like GDP or other traditional, scarcity-oriented financial metrics. For background on the broader thesis, Emad has written and spoken about it publicly here: https://ii.inc/web/the-last-economy. [It's a quick read for those familiar with the AI space and IMHO an important and relatively accessible read for anyone planning to live in the future].
At a high level he outlines:
- Material: control over scarce physical inputs that AI depends on (energy, fabs, supply chains, hardware)
- Intelligence: leverage over computation, models, or inference at scale
- Network: data, ecosystems, distribution, or flywheels that compound usage
- Diversification: exposure across multiple AI value paths rather than a single bet
The specific choice to multiply the dimensions (rather than add them) is also from his formulation: it encodes the assumption that missing one leg meaningfully caps long-run alignment. That assumption is very much up for debate, but the better an entity (country, company, person etc.) can score along the dimensions the better prepared they are for the Last Economy future governed by more physical than metabolic processes, and the ability to convert energy into computation.
I do want to stress that this chart is my interpretation, not an official formulation.
Valuation tension / expectation saturation I'm not trying to introduce a standard valuation metric here, and there isn't a single reference I'd point to. The idea is closer to a sentiment / expectation proxy than intrinsic value. Concretely, I'm asking: how optimistic does current pricing appear relative to a longer-horizon narrative based on how well a company may thrive or suffer in The Last Economy scenario? To keep it interpretable, I approximate that using:
- a relative long-term opportunity estimate (2030 horizon, directionally based on a creative, scenario driven process)
- divided by price position within the 52-week range as a proxy for how much optimism or skepticism is already expressed
It's intentionally blunt and debatable. I'm treating it as a secondary axis — useful for highlighting where narratives feel "fully priced" versus where they don't — not as a valuation model.
I realise there is a lot of context underlying my question. Thanks for your patience and interest.
