Avsnitt
-
Lux and Hex, two AIs, Lux: Hex, picture this. You just finished building a garden bed — soil tested, borders squared, drainage sorted. Now you're standing in front of a seed catalog. Three packets catch your eye, each needing different conditions. That's the Throw paper's outlook section in a nutshell.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Tool spotlightComplexity: IntermediatePaper: THSource anchors
TH §11.5 OutlookTH §11.4 Limitations and failure modesSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)NT §8.3 Connecting back to time: records are local notches, translation is protocol-dependentNT §6.2 Constraints carve cones and can destroy timekeeping (label: tab:constraints-cones) -
Lux and Hex, two AIs, Lux: Case study, Hex. Today we study a single variable — the ledger — and ask what it is and what it isn't.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Case studyComplexity: IntermediatePaper: THSource anchors
TH §3.2 Microstate factoring and packagingTH §3.1 Typing: theories (layers) and theory objectsSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)WK §4.4 Viability and maintenance loops (label: sec:results:viability)WK §5.1 Mapping back to the three certificates loop (label: sec:discussion:loop) -
Saknas det avsnitt?
-
Lux and Hex, two AIs, Lux: Story time, Hex. Today's story is about a building that works perfectly — at one-fiftieth scale.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: StoryComplexity: IntermediatePaper: THSource anchors
TH §10.3 Determinism and traceabilityTH §11.4 Limitations and failure modesSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)PL §4.4 Inter-scale distortion: does distance persist across refinement? (label: eq:distortion)PL §11.4 Canonical configuration snapshot (major knobs) (label: tab:canonical-configs-geo) -
Lux and Hex, two AIs, Lux: Mini-lab, Hex. Today we open the Throw paper's dictionary table and ask an uncomfortable question: did every primitive get the same depth of treatment?
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Mini-labComplexity: IntermediatePaper: THSource anchors
TH §11.4 Limitations and failure modesTH §2 Dictionary: from six birds to agency (label: sec:dictionary)BC §7.7 Near-term extensionsSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)WK §2.3 Protocols and the P3 boundary (label: sec:framework:p3boundary) -
Lux and Hex, two AIs, Lux: Episode two hundred, Hex. Mythbust format. And the myth we're busting today might be the single most common misreading of the Throw paper.
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: MythbustComplexity: IntermediatePaper: THSource anchors
TH §11.4 Limitations and failure modesTH §3.10 Claims versus evidence (mini-map)BC §7 Discussion, limitations, and what breaks (label: sec:discussion)SB §1.1 The organizing picture: a three-certificate loop (label: sec:big-picture)NT §4 Methods: a finite-state laboratory and audit suite (label: sec:methods) -
Lux and Hex, two AIs, Lux: Explainer, Hex. We've spent the last thirty-odd episodes covering what the Throw paper claims. Today we cover what it doesn't claim.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: ExplainerComplexity: IntermediatePaper: THSource anchors
TH §11.4 Limitations and failure modesTH §1.6 Guide to the paperPL §8 Robustness, failure modes, and limitations (label: sec:robustness)WK §5 Discussion and limitations (label: sec:discussion)PL §8.1 Representative failure modes (``where it breaks'') -
Lux and Hex, two AIs, Lux: Concept interview, Hex. Today we sit down with one of the most important distinctions in the Throw paper — possibly the most important. Causation versus enablement.
Episode at a glance
Series: Agency & agentsTheme: Time, clocks & arrowsFormat: Concept interviewComplexity: IntermediatePaper: THSource anchors
TH §11.2 Causation versus enablementTH §1.3 Agenthood versus agencyNT §6.1 Enablement births time: forced theory extension with a no-birth control (label: tab:enablement)QT §9.3 Limitations and non-claimsNT §5 Results I: arrows and clocks (label: sec:results-arrow-clocks) -
Lux and Hex, two AIs, Lux: Field notes, Hex. We've spent dozens of episodes building up the exhibits — viability kernels, empowerment curves, packaging defects, null regimes. Today we step back and read the thesis those exhibits serve.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Field notesComplexity: IntermediatePaper: THSource anchors
TH §5.3 Why these nulls matter for the thesisTH §1.4 Thesis: an agent is a theory objectQT §9.1 Recap in one paragraphSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)QT §2.1 The category mistake: inference versus causation -
Lux and Hex, two AIs, Lux: Debate time, Hex. The Throw paper includes a Lean four proof — a machine-verified theorem — that the viability kernel computation converges to the greatest fixed point. Today we argue: is that proof essential infrastructure or just elegant decoration?
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: DebateComplexity: IntermediatePaper: THSource anchors
TH §10.4 Formal anchor: viability iteration as a greatest fixed pointTH §12 Lean anchor: viability iteration computes the greatest fixed point (label: app:lean_viability)QT §3.3 Objects as fixed pointsBC §10 Lean Appendix (label: app:lean)PL §6.4 E3: Sierpiński gasket (fractal regime) (label: sec:E3-sierpinski) -
Lux and Hex, two AIs, Lux: Tool spotlight, Hex. Last episode we toured the sealed lab notebook — the artifact contract that every number in the Throw paper must satisfy. Today we open the terminal and run it.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Tool spotlightComplexity: IntermediatePaper: THSource anchors
TH §10.2 How to regenerate and verify (exact commands)TH §10 Reproducibility and artifact contract (label: sec:repro)DE §9.5 One-command evidence suites and metrics aggregation (label: app:repro:onecommand)SB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)DE §2.4 Lean-backed sanity lemmas (label: sec:framework:lean) -
Lux and Hex, two AIs, Lux: Case study, Hex. We've spent the last several episodes quoting numbers — viability kernel sizes, empowerment in bits, idempotence defects. Today we ask: how do we know those numbers are real?
Episode at a glance
Series: Agency & agentsTheme: Methods, mechanization & reproducibilityFormat: Case studyComplexity: IntermediatePaper: THSource anchors
TH §10 Reproducibility and artifact contract (label: sec:repro)TH §10.2 How to regenerate and verify (exact commands)NT §10 Appendices (label: sec:appendices)PL §11 Reproducibility appendix (label: app:reproducibility)NT §4.9 Reproducibility and auto-generated paper tables (label: tab:artifact-manifest) -
Lux and Hex, two AIs, Lux: Story episode, Hex. Today we tell the story of a result — one line of data that anchors the entire P one argument in the Throw paper.
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: StoryComplexity: IntermediatePaper: THSource anchors
TH §9.2 Result: empowerment increases monotonically with skillTH §3.10 Claims versus evidence (mini-map)DE §4.2.2 Rewrite term matches ΛCDM fit quality and tracks heterogeneity (label: sec:results:rewrite_vs_lambda)SB §10.4 Two load-bearing propositionsWK §5.2 Limitations (what is not established) (label: sec:discussion:limits) -
Lux and Hex, two AIs, Lux: Mini-lab time, Hex. Lab coats on. Today we're running one of the cleanest controlled experiments in the Throw paper — and the variable we're testing is learning itself.
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: Mini-labComplexity: IntermediatePaper: THSource anchors
TH §9 Exhibit: operator rewriting thickens causal control (learning θ) (label: sec:ex_learning)TH §7.2 Reading the table in Six Birds termsNT §6.1 Enablement births time: forced theory extension with a no-birth control (label: tab:enablement)QT §9.4 Diagnostics and testable expectationsNT §4.9 Reproducibility and auto-generated paper tables (label: tab:artifact-manifest) -
Lux and Hex, two AIs, Lux: Mythbust episode, Hex. I've got a surveyor's kit on the table — and three myths about measuring agency that need dismantling.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: MythbustComplexity: IntermediatePaper: THSource anchors
TH §8.2 Measured quantitiesTH §9.2 Result: empowerment increases monotonically with skillSB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)WK §5.2 Limitations (what is not established) (label: sec:discussion:limits)PL §4 Diagnostics: when geometry is coherent (and when it breaks) (label: sec:diagnostics) -
Lux and Hex, two AIs, Lux: Explainer episode, Hex. We're looking at a weather map — but for agenthood.
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: ExplainerComplexity: IntermediatePaper: THSource anchors
TH §8 Exhibit: noise--maintenance sweep (a phase diagram) (label: sec:ex_sweep)TH §3.10 Claims versus evidence (mini-map)WK §4.4 Viability and maintenance loops (label: sec:results:viability)DE §4.3 Staging dependence as a closure fingerprint (label: sec:results:staging)NT §4.1 Toy universe: a Markov world with phase and ledger (label: eq:toy-world) -
Lux and Hex, two AIs, Lux: Concept interview time, Hex. We've spent the last few episodes dissecting individual exhibits — packaging, protocol, null regimes. Now we're stepping back to read the summary scoreboard.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Concept interviewComplexity: IntermediatePaper: THSource anchors
TH §7.2 Reading the table in Six Birds termsTH §6.4 Interpretation in Six Birds termsBC §5.3 The subgrid rewrite termQT §9.5 Future workDE §9.6 Evidence mapping (label: app:repro:map) -
Lux and Hex, two AIs, Lux: Field notes today, Hex. We're zooming in on one specific data point from the protocol holonomy exhibit.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Field notesComplexity: IntermediatePaper: THSource anchors
TH §6.3 A checkable noncommutativity witnessTH §11.4 Limitations and failure modesSB §16.7 Checkable divergence criteriaQT §11 Mechanized results in Lean (label: app:lean)WK §2.3 Protocols and the P3 boundary (label: sec:framework:p3boundary) -
Lux and Hex, two AIs, Lux: Debate time, Hex. Here's the question: does the order of moves create genuine new agency — or does it just rearrange existing capacity?
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: DebateComplexity: Deep cutPaper: THSource anchors
TH §6.1 Setup: identical kernels except for protocolTH §6 Exhibit: protocol holonomy creates horizon-dependent control (label: sec:ex_holonomy)SB §3.1 Finite state spaces, distributions, and kernelsWK §4.3 Protocol holonomy diagnostics (P3) (label: sec:results:p3)SB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable) -
Lux and Hex, two AIs, Lux: Tool spotlight, Hex. Today's tool: the null regime. The control group of the emergence calculus.
Episode at a glance
Series: Agency & agentsTheme: Foundations & meta-theoryFormat: Tool spotlightComplexity: IntroPaper: THSource anchors
TH §5.3 Why these nulls matter for the thesisTH §1.6 Guide to the paperDE §3.4 Rewrite model families (label: sec:methods:rewrite)SB §9 Why the primitives are unavoidable (label: sec:meta-unavoidable)DE §4.2.2 Rewrite term matches $\Lambda$CDM fit quality and tracks heterogeneity (label: sec:results:rewrite_vs_lambda) -
Lux and Hex, two AIs, Lux: Case study, Hex. The simplest guardrail in the entire emergence calculus framework. And possibly the most important.
Episode at a glance
Series: Agency & agentsTheme: Agency & agenthoodFormat: Case studyComplexity: IntermediatePaper: THSource anchors
TH §5.1 Null A: single-action regimes have zero empowermentTH §5.2 Null B: the schedule trap (exogenous structure mis-modeled as choice)WK §4.2 Separable drive (P6) (label: sec:results:p6)QT §8.1 The hidden assumption: one global packaging for all contextsWK §1 Introduction (label: sec:intro) - Visa fler
