https://sourcescore.org/blog/ Posts on AI-citation quality, LLM grounding, and the SourceScore methodology. en-US Sat, 16 May 2026 17:47:02 GMT scripts/generate-rss.mjs https://sourcescore.org/blog/why-no-performance-claims/ https://sourcescore.org/blog/why-no-performance-claims/ Sat, 16 May 2026 00:00:00 GMT Benchmark numbers vary by prompt format, model version, shot count, and evaluation harness. Shipping them as verified claims is the surest way to make the catalog wrong by Thursday. The methodology rules we keep — and the trust math they preserve. methodology, trust, benchmarks, veritas https://sourcescore.org/blog/verify-ai-facts-five-lines-python/ https://sourcescore.org/blog/verify-ai-facts-five-lines-python/ Sat, 16 May 2026 00:00:00 GMT Drop SourceScore VERITAS into your LLM pipeline as a post-generation check. Every claim the model emits gets a confidence score plus canonical citation before the user sees it. Five-line client + a generate-then-verify loop pattern. tutorial, python, veritas, hallucination https://sourcescore.org/blog/launching-veritas/ https://sourcescore.org/blog/launching-veritas/ Sat, 16 May 2026 00:00:00 GMT VERITAS is a free-tier-friendly API that returns hand-verified AI/ML claims with their primary sources, an HMAC-SHA256 signature, and a ready-to-paste citation. 100 claims today, expanding through Q3. launch, veritas, api, llm-grounding