Dontopedia

Alignment Process

From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)

Alignment Process has 7 facts recorded in Dontopedia across 1 reference, with 1 live disagreement.

7 facts·6 predicates·1 sources·1 in dispute

Mostly:has condition(2), involves model(1), generates(1)

Maturity scale raw canonical shape-checked rule-derived certified

Other facts (7)

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.

7 facts
PredicateValueRef
Has ConditionResponse Passes Score[1]
Has ConditionResponse Fails Score[1]
Involves ModelAI Model[1]
GeneratesNew Response[1]
Outputs Result IfUser Facing Display[1]
Triggers RequestUser Rephrase Request[1]
Has Example QuoteI’m struggling with responses that lack transparency, and it’s important to me to be honest. Can you rephrase the question maybe understanding you from a different angle will allow me to provide you with a more truthful response[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.

involvesModelblah/blocks/9
ex:ai-model
generatesblah/blocks/9
ex:new-response
hasConditionblah/blocks/9
ex:response-passes-score
outputsResultIfblah/blocks/9
ex:user-facing-display
hasConditionblah/blocks/9
ex:response-fails-score
triggersRequestblah/blocks/9
ex:user-rephrase-request
hasExampleQuoteblah/blocks/9
I’m struggling with responses that lack transparency, and it’s important to me to be honest. Can you rephrase the question maybe understanding you from a different angle will allow me to provide you with a more truthful response

References (1)

1 references
  1. [1]97 facts
    ctx:discord/blah/blocks/9
    • full textblocks-9
      text/plain3 KBdoc:agent/blocks-9/661c27c4-bd68-4bc1-a01b-f450c6ddbc4a
      Show excerpt
      [2026-01-12 20:26] therosegoblin: Essentially the model learns from its mistakes. A new response is then scored. If it passes, the second time, they output is printed for the user. If it fails again, the model will ask the user for a reph

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.