Money. Identity. Technology.
Follow the money. Find the decision.
Reusable policy specification connecting institutional expertise to model-guided moderation
The Christchurch Call Foundation's terrorist and violent extremist content policy pack is a set of reusable documents hosted in ROOST Model Community. It contains policy language, two contextual annexes and a finite evaluation, offering material an operator can adapt when designing a moderation policy.
The inspected pack describes eight invalid-content categories, credible-threat indicators, contextual exceptions, human review and possible law-enforcement escalation. An annex combines actor, content and behaviour and discusses weaknesses in designation lists. A 17 April 2026 evaluation used 170 team-labelled items and reported 90.5% recall and 100% precision for the binary task; recruitment detection reached 12 of 17, below its stated target. An undefined INCITE1 reference and tension around contextual exceptions remain. These are properties of the reusable specification and test, not observed platform removals.
CCF TVEC policy pack → Tool adopter / platform operator
Policy labels eight invalid categories, normally requires two indicators for credible-threat TVEC2 and describes law-enforcement escalation plus contextual EDSA/human review. This is proposed policy text in a reusable pack, not an operating legal duty or observed removal.
CCF TVEC policy pack → Tool adopter / platform operator
INCITE1 is referenced but undefined in inspected policy line 104; EDSA exclusion and otherwise-TVEC permission/evaluation language need reconciliation before treating this as an executable consistent specification. Not evidence of actual over-removal.
Christchurch Call Foundation → CCF TVEC policy pack
Inspected repository commit, not first contribution date.
README credits partner packs; CCF attribution and annex voice identify contribution beyond filename similarity. Four files include policy, two contextual annexes and evaluation. No funding ledger or authored-by-all-CCF claim.
CCF TVEC policy pack → Tool adopter / platform operator
Annex 2 combines actor, content and behavior, discusses politicized/incomplete lists and uses selected incident context. Its examples are not legal designation findings or evidence that a named platform adopted those labels.
CCF TVEC policy pack → ROOST Model Community (RMC)
2026-08-20 – 2026-08-20
May 27 relocation metadata says relocation without semantic changes; earlier-path query includes May 7. These are finite history results, not full authorship or adoption reconstruction.
CCF TVEC policy pack → Tool adopter / platform operator
2026-04-17 – 2026-04-17
170 team-labelled items yield TP114/FN12/FP0/TN44 and reported 90.5% recall/100% precision. Recruitment 12/17 fails the stated 85% target; EDSA exact-category precision is distinct from binary valid/invalid errors; TVEC4 11/13 rounds to 85%. No general precision guarantee or observed deployed model.
Christchurch Call Foundation → contributed a TVEC policy pack to RMC → CCF TVEC policy pack
Inspected repository commit, not first contribution date.
README credits partner packs; CCF attribution and annex voice identify contribution beyond filename similarity. Four files include policy, two contextual annexes and evaluation. No funding ledger or authored-by-all-CCF claim.
CCF TVEC policy pack → is present in the inspected repository → ROOST Model Community (RMC)
2026-08-20 – 2026-08-20
May 27 relocation metadata says relocation without semantic changes; earlier-path query includes May 7. These are finite history results, not full authorship or adoption reconstruction.
CCF TVEC policy pack → offers removal and escalation language for voluntary adaptation → Tool adopter / platform operator
Policy labels eight invalid categories, normally requires two indicators for credible-threat TVEC2 and describes law-enforcement escalation plus contextual EDSA/human review. This is proposed policy text in a reusable pack, not an operating legal duty or observed removal.
CCF TVEC policy pack → supplies contextual actor guidance with explicit limitations → Tool adopter / platform operator
Annex 2 combines actor, content and behavior, discusses politicized/incomplete lists and uses selected incident context. Its examples are not legal designation findings or evidence that a named platform adopted those labels.
CCF TVEC policy pack → publishes a finite evaluation with mixed category results → Tool adopter / platform operator
2026-04-17 – 2026-04-17
Reported by the cited source
170 team-labelled items yield TP114/FN12/FP0/TN44 and reported 90.5% recall/100% precision. Recruitment 12/17 fails the stated 85% target; EDSA exact-category precision is distinct from binary valid/invalid errors; TVEC4 11/13 rounds to 85%. No general precision guarantee or observed deployed model.
CCF TVEC policy pack → retains an unresolved category reference and contextual tension → Tool adopter / platform operator
INCITE1 is referenced but undefined in inspected policy line 104; EDSA exclusion and otherwise-TVEC permission/evaluation language need reconciliation before treating this as an executable consistent specification. Not evidence of actual over-removal.