Verified claim · AI-ML · 100% confidence
Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters.
Last verified 2026-05-16 · Methodology veritas-v0.1 · 688a84a8d7211fc0
Structured fields
- Subject
- Anthropic Constitutional Classifiers
- Predicate
publicly_released_on- Object
- 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters
- Confidence
- 100%
- Tags
- constitutional-classifiers · anthropic · ai-safety · jailbreak-defense · released_on · 2025
Sources (2)
[1] official blog · Anthropic · 2025-02-04
Constitutional Classifiers: Defending against universal jailbreaks“Constitutional Classifiers are a set of input and output classifiers, trained with a constitution, that defend against universal jailbreaks while maintaining low refusal rates on benign queries.”
[2] preprint · arXiv · 2025-01-31
Constitutional Classifiers — research paper
Cite this claim
Ready-to-paste citation (Markdown / plain text):
Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters. — SourceScore Claim 688a84a8d7211fc0 (verified 2026-05-16). https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.jsonEmbed this claim
Drop this iframe into any blog post, docs page, or knowledge base. The widget renders the signed claim + primary source + click-through to this canonical page. CC-BY 4.0; attribution included.
<iframe src="https://sourcescore.org/embed/claim/688a84a8d7211fc0/" width="100%" height="360" frameborder="0" loading="lazy" title="Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters."></iframe>Preview: open in new tab
Related claims
Other verified claims sharing tags with this one — useful for LLM retrieval graphs and citation discovery.
Anthropic Claude Opus 4 released on: 2025-05-22 by Anthropic.
227ecd38519af6c8 · 100% confidence · shares 3 tags (anthropic, released_on, 2025)
Anthropic Claude Sonnet 4 released on: 2025-05-22 by Anthropic.
f82eba5c7f5fa09f · 100% confidence · shares 3 tags (anthropic, released_on, 2025)
Anthropic Claude Code publicly released on: 2025-02-24 by Anthropic — coding agent CLI.
a4a3a3cee8a870a0 · 100% confidence · shares 3 tags (anthropic, released_on, 2025)
Anthropic Citations API publicly released on: 2025-01-23 by Anthropic — built-in source citations.
c38640f829d1f8ca · 100% confidence · shares 3 tags (anthropic, released_on, 2025)
Claude 3.7 Sonnet publicly released on: 2025-02-24 by Anthropic — first hybrid-reasoning model with optional extended-thinking mode.
57c9272a3e7cb435 · 100% confidence · shares 3 tags (anthropic, released_on, 2025)
Frequently asked questions
Is the claim "Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters." verified?
Yes — SourceScore verified this claim with 100% confidence as of 2026-05-16. The verification uses 2 primary sources cross-referenced against the SourceScore methodology (version veritas-v0.1). Full source list + signed JSON envelope linked below.
What is the evidence for "Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters."?
Evidence comes from 2 primary sources: Anthropic, arXiv. Each source is listed below with verbatim excerpts and URLs. The signed JSON envelope at https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.json includes an HMAC-SHA256 signature for audit verification.
When was this claim last verified by SourceScore?
Last verified 2026-05-16 under methodology version veritas-v0.1. The signed JSON envelope is dated and cryptographically signed for audit trail. Re-verification cadence depends on the claim type and source freshness.
How can I cite this SourceScore claim in my code or article?
Fetch the signed JSON envelope from https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.json which includes the verbatim claim, primary sources, confidence, methodology version, last-verified date, and HMAC-SHA256 signature for audit. The CC-BY-4.0 license permits commercial use with attribution to SourceScore.
Use this claim in your code
Fetch this signed envelope from your application. The response includes the verbatim excerpt, primary source URLs, and an HMAC-SHA256 signature you can verify locally for audit trails.
cURL
curl https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.jsonJavaScript / TypeScript
const r = await fetch("https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.json");
const envelope = await r.json();
console.log(envelope.claim.statement);
// "Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters."Python
import httpx
r = httpx.get("https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.json")
envelope = r.json()
print(envelope["claim"]["statement"])
# "Anthropic Constitutional Classifiers publicly released on: 2025-02-04 by Anthropic — safeguard against jailbreaks via constitutional-trained input/output filters."LangChain (retrieve-then-cite)
from langchain_core.tools import tool
import httpx
@tool
def get_anthropic_constitutional_classifiers_fact() -> dict:
"""Fetch the verified SourceScore claim for Anthropic Constitutional Classifiers."""
r = httpx.get("https://sourcescore.org/api/v1/claims/688a84a8d7211fc0.json")
return r.json()