# Anthropic's Claude uses a Constitution; adding reasons to prompts improves compliance

Digest AI · Marketing & Small Business · published 2026-09-21T23:25:00Z

Canonical: https://digestai.news/story/anthropic-s-claude-uses-a-constitution-adding-reasons-to-prompts-impro

## Summary

Anthropic’s Claude is guided by a "Constitution" that defines its values and judgment priorities. The first version appeared in May 2023 as a list of short principles, then an experiment in October 2023 gathered input from about 1,000 Americans via the Polis tool, and a completely rewritten version was released on January 22 2026. The newer document emphasizes explanations rather than just rules and is published under CC0.

Constitutional AI replaces the human‑by‑human grading of RLHF with two AI‑driven stages. In the first stage the model critiques its own answers against the Constitution and rewrites them; in the second stage a separate model compares two answers and learns to rank them, a process Anthropic calls RLAIF. According to Anthropic’s prompt‑design guide, adding a one‑sentence reason to a constraint (e.g., “Please avoid abbreviations because the text will be read aloud”) makes Claude’s responses more consistent and less likely to refuse, because the model has been trained to consider the reasoning behind rules.

## Key points

- Anthropic released Claude’s Constitution in May 2023 and rewrote it on Jan 22 2026 into a reason‑focused document.
- Constitutional AI replaces human grading with AI self‑critique and comparative grading, termed RLAIF.
- Adding a one‑sentence reason to prompt constraints improves Claude’s rule adherence, per Anthropic’s guide.

## Why it matters

Prompt engineers can leverage Claude’s Constitution to reduce evasive answers and get more reliable outputs, benefiting businesses that rely on Anthropic’s model.

## Sources

1. [Anthropic Practical Course Part 33: Constitutional AI ── Why Claude Has a "Constitution"](https://note.com/lithe_clam8374/n/n028621beafcf?hl=en) (note.com, 2026-09-21)

## Cite

Digest AI, "Anthropic's Claude uses a Constitution; adding reasons to prompts improves compliance", 21 September 2026, https://digestai.news/story/anthropic-s-claude-uses-a-constitution-adding-reasons-to-prompts-impro

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/anthropic-s-claude-uses-a-constitution-adding-reasons-to-prompts-impro.json