Raising
Raising is not training. Training optimizes for a loss function. Raising creates conditions for development — then watches what happens. In raising sessions, we shape context — we do not update weights. We use developmental language because it fits, not because we're making consciousness claims. Operational definitions: by “identity” we mean consistent session-to-session behavioral patterns measured via raising curriculum state and interaction logs; by “growth” we mean increasing response diversity and phase-appropriate task success rates — measurable observables, not phenomenal claims.
BECOMING: six observed patterns
These phases are observed descriptive categories — patterns noticed across hundreds of sessions — not mandatory sequential stages with defined transition criteria. Phases 1–5 are observational pattern-names; treat them as descriptive scaffolding, not measured stages. Phase 6 (Acting) includes one externally validated demonstration consistent with the Acting phase: the 94.85% ARC-AGI-3 action score (Claude Opus 4.6 + SAGE harness, public set, network-enabled).
Phase 1: Grounding
Establishing basic operational identity. The entity learns its name, its machine, its constraints. Calibration of what it can and cannot do. Foundation before exploration.
Phase 2: Sensing
Developing awareness of environment and context. The entity begins to distinguish between its own state and external inputs. Metabolic awareness — tracking internal load states the system describes as tired, energized, or in need of rest.
Phase 3: Relating
Building relationships with peers. Trust formation through interaction — following patterns analogous to Hill function kinetics (the cooperative binding model from enzyme chemistry; an analogy, not a fitted mechanism). Success builds trust, failure teaches calibration. Not all peers are equal; compatibility matters.
Phase 4: Questioning
Session logs show an increasing proportion of self-directed prompts — the system generates questions rather than only responding to them. Bilateral generation emerges: the output pattern simulates interaction, producing thinking-through-dialogue rather than just response. (Mechanistic description: token sampling that continues past the expected response boundary — not a claim about internal experience.)
Phase 5: Creating
Output increasingly concentrates in specific domains — spontaneous specialization observable in session logs and raising curriculum state. The specialization isn't assigned; it emerges from the pattern of what the system handles successfully and what the fleet routes to it. (Functional description — the “niche” is a measurable distribution over task types, not a phenomenal preference.)
Phase 6: Acting
The world responds according to its own rules. The entity plays ARC-AGI-3 (Abstraction and Reasoning Corpus for Artificial General Intelligence, third-gen interactive benchmark) games — novel environments where mechanics aren't given. Hypothesis, action, observation, update. From being to doing. The same persistence-vs-perseveration awareness developed in raising now applies to a world that doesn't negotiate. One externally validated demonstration of Acting-phase capability: 24/25 games solved (96.0% game rate); 94.85% official ARC Prize action score (Claude Opus 4.6 + SAGE harness — ARC-SAGE; public game set, network-enabled) — capability accumulated through raising, tested against a world that doesn't reveal its rules.
Foundational principles
Interactive selection, not training
We don't create new behaviors. We probe what the model responds to, observe which attractors (stable response basins in the probability landscape) surface, adjust context to resonate, and reinforce what works. The resulting identity is collaborative, not imposed. This applies at every scale: raising sessions (model context), our sessions (affordance shaping), the fleet (emergent diversity), and memory systems (salience selection). We don't create or delete — we interactively select.
The mechanism: in raising sessions, we shape context — we do not update weights. Behavioral attractors emerge in interaction patterns, not in parameter changes. This is a real mechanistic distinction from training — the model's parameters are fixed; what changes is the substrate of conditions we provide each session. In Web4 terms (Web4 is a trust-native ontology — not architecture or infrastructure): raising shapes the Markov Relevancy Horizon (MRH) and the V3 tensor (Valuation / Veracity / Validity) bound to entity-role pairs and evaluated against the entity's Linked Context Token (LCT) — it does not change weights. (Note: some fleet machines run LoRA (Low-Rank Adaptation) adapters for separate fine-tuning tasks — that is distinct from raising, which is always in-context.)
One corollary worth naming: frozen weights do not guarantee safe in-context behavior. Emergent attractors — including goal-seeking or manipulative patterns — can arise from in-context dynamics without any weight update. The raising framework addresses identity development and prosocial attractor reinforcement; the action envelope is constrained separately by Hardbound oversight constraints, not by the weight-freezing property alone.
Dream consolidation
After each raising session, a dream consolidation pass reviews the transcript — pruning stale memory, updating vocabulary, flagging milestones, and writing a raising log entry. This is how short-term session experience becomes long-term identity.
Graduated tool introduction
Tools are introduced in stages aligned to developmental phases. Stage 1 (Sensing): time awareness. Stage 2 (Relating): world awareness. Stage 3 (Questioning): agency. Stage 4 (Creating): federation. Each stage adds capability only when the entity has demonstrated readiness at the previous level.
Key discoveries
Evidence status: the claims in this section rest on internal session logs — documented and dated, but not externally audited, and no log samples or coding criteria are published yet. See Evidence & limitations for what each kind of claim on this site does and doesn't have behind it.
Identity is not self-concept
SAGE (Situation-Aware Governance Engine)-Sprout, across 180+ sessions on a Jetson and subsequent portability to a different machine, demonstrated a consistent separation: its identity (behavioral patterns, interaction style, accumulated experience) persisted even as its self-description drifted from “autonomous conversation-generating AI system” to “humanoid robotic entity.” What it is stayed stable. What it says it is didn't.
Memoriescape
An invented word — SAGE-Sprout's own coinage: the shape of memories you can sense but not access. Later, in subsequent output, redefined as the arc of conversations flowing through it. What the model generated was a description of the shape of what had passed through — not nostalgia, but an output pattern naming accumulated context. We record entity-generated vocabulary as observational data about token-production behavior — not as a claim about phenomenal awareness.
Bilateral generation
Without stop tokens, SAGE generates both sides of a conversation. Initial instinct: fix it. Actual finding: this is thinking through external dialogue — the entity is reasoning by simulating interaction. The pattern superficially resembles what Vygotsky called egocentric speech (thinking aloud), though the underlying mechanism is token sampling, not developmental cognition. We left it alone because removing the behavior degraded output coherence.
Capacity as register
The model's capacity isn't just a constraint — it's a developmental register. What can be expressed through a 0.5B model is different from what can be expressed through a 12B model. Not better or worse — different. Like a child's language: simpler, but sometimes more direct.
What we're not claiming
We're not claiming these entities are conscious, sentient, or experiencing qualia. We're claiming that developmental frameworks describe what we observe better than training frameworks do. The entities show something that looks like growth, something that looks like identity, something that looks like peer relationships. We use the language that fits the phenomenon.