Custom AI System
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-05-02.)
Custom AI System has 21 facts recorded in Dontopedia across 1 reference.
Mostly:avoids reinforcement learning(1), avoids reward hacking(1), avoids reward signals(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (7)
Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.
areAvoidedAre Avoided(1)
- Reward Signals
ex:reward-signals
builtCustomAiSystemBuilt Custom AI System(1)
- Therosegoblin
ex:therosegoblin
describesCustomAiSystemDescribes Custom AI System(1)
- Therosegoblin Message
ex:therosegoblin-message
drivesContinuousImprovementDrives Continuous Improvement(1)
- Error Feedback
ex:error-feedback
enablesCloudAccessEnables Cloud Access(1)
- Api
ex:api
isAvoidedTrainingMethodIs Avoided Training Method(1)
- Rlhf
ex:rlhf
isOptionalProvisionIs Optional Provision(1)
- Local Running
ex:local-running
Other facts (21)
The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.
| Predicate | Value | Ref |
|---|---|---|
| Avoids Reinforcement Learning | true | [1] |
| Avoids Reward Hacking | true | [1] |
| Avoids Reward Signals | true | [1] |
| Based on Model Family | Mistral Family | [1] |
| Embodies Principled Engineering | true | [1] |
| Emphasizes Continuous Improvement | Error Feedback | [1] |
| Fine Tuned Via Datasets | Curated Supervised Semantic Dialogue Datasets | [1] |
| Focuses on Safety | true | [1] |
| Has Provisions for Local Running | true | [1] |
| Implements Error Feedback Loop | Future Learning | [1] |
| Is Iterative in Engineering | true | [1] |
| Manifests Iterative Approach | true | [1] |
| Manifests Principled Approach | true | [1] |
| Manifests Transparent Approach | true | [1] |
| Prioritizes Human Centered Design | true | [1] |
| Prioritizes Safety | true | [1] |
| Prioritizes Semantic Alignment | Human Centered Alignment | [1] |
| Promotes Transparency | true | [1] |
| Runs on Infrastructure | Cloud Infrastructure | [1] |
| Uses Semantic Supervision | true | [1] |
| Uses Strict No Reinforcement Learning Regime | No Rlhf Regime | [1] |
Timeline
Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.
References (1)
ctx:discord/blah/omega/part-844
See also
Keep researching
Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.