I design fun, engaging, insight-rich sessions that help teams move from curiosity and skepticism to excitement-driven alignment with concrete next steps.
Most AI adoption efforts stall for a simple reason:
People are shown AI.
They are not meaningfully engaged with it.
Someone demos something impressive.
Everyone nods.
Then they go back to work and nothing changes.
Why?
1
@Sirlupinwatson I'd think its more local to what you are local to, not impossible. And what your local to is what the other observers in a subsystem with some constraints are confirming. (nearby, city, continent, planet, solar system, galaxy....)
We are observers.
Observers possess:
- local records
- limited observation capability
- a model for reconstructing the world
The world state itself is not directly accessible.
Explore mathematical frameworks in which observers maintain partial records of a shared causal history, and reality corresponds not to a single globally accessible state but to the set of invariants that remain stable across admissible reconstructions from distributed records.
We’ve added a CLI for Claude Platform to make every API endpoint runnable from your terminal.
Call the Messages API, stand up Claude Managed Agents, pipe results straight into your shell.
The ant CLI is well understood by coding agents (Claude Code) using the claude-api skill.
It finally arrived!
Will dig into @tjayrush essays tonight.
I love seeing the thoughts of people I like and admire put to print.
The user owns the data or the whole thing was pointless.
@8ctopuso https://t.co/HzqPdQClCE
I wanted to create a few research repos interesting enough to try and prove out a protocol design. Didn’t get to that stage yet… hobby speed
@EnesAvar Admissible = consistent with the accumulated constraints. The more records, observations, and finalized history something must satisfy, the fewer renderings remain possible.
Can you formalize a toy model where observers are partial record-keepers in a graph-of-graphs, “visible matter” = finalized structure renderable from one region, and “dark” structure = finalized but not locally renderable? What assumptions could yield or falsify a ~5:1 hidden/visible ratio?
Each observer reconstructs a rendering from the constraints imposed by available records.
How does the space of admissible renderings shrink as records accumulate and propagate?
Shared reality hardens where observer-record paths overlap. Matter is experienced as stable constraint on possible transformation, not as mere agreement.
A. Other minds as indirectly certified via ZK/proof-carrying traces.
B. Matter/body as constraints/invariants, not agreement.
C. Body as boundary/interface via tensor-network style thinking.
D. Observer-renderer as accessible observables/subalgebra.
E. Shared reality as overlap-hardening across causal DAGs.
A conscious observer renders an action-guiding world from accessible records and constraints. The body is the high-finality interface through which the observer becomes causally addressable. Other minds are independently real but indirectly certified.