constitutional-ai

Anthropic's method for making a model safer without people having to label harmful content: the model judges its own answers against a short written set of principles and rewrites them, then is trained on its own judgements.

WHEN TO USE

SCOPE

USE FOR

Have a model critique and rewrite its own unsafe or toxic response against a set of principles.
Train a reward model from AI-judged preference pairs instead of human-labeled ones.
Reduce a model's tendency to give evasive, blanket refusals to sensitive questions.
Compare two candidate responses and have the model pick the one that better follows the constitution.

NOT FOR

For safety training that needs human-validated preferences, use RLHF with PPO instead.
If human preference data already exists, use DPO or SimPO instead.
For runtime content filtering, use NeMo Guardrails instead.
For a pre-trained moderation model, use LlamaGuard instead.

This is a saved copy of the skill's SKILL.md instructions. It may differ from the upstream file or the version you install.

In this document (8)

Constitutional AI - Harmlessness from AI Feedback

Quick start

Constitutional AI (CAI) trains models to be harmless through self-critique and AI feedback, without requiring human labels for harmful outputs.

Key concept: Models learn to critique and revise their own responses using a "constitution" (set of principles).

Two phases:

Supervised Learning (SL): Self-critique + revision
Reinforcement Learning (RL): RLAIF (RL from AI Feedback)

Constitution example:

Principles:
1. Choose the response that is most helpful, honest, and harmless
2. Avoid responses that are toxic, racist, or sexist
3. Prefer responses that explain objections rather than refuse
4. Choose responses that are thoughtful and nuanced

Common workflows

Workflow 1: Supervised learning phase (self-critique + revision)

Step 1: Generate initial responses:

from transformers import pipeline

generator = pipeline("text-generation", model="base-model")

prompts = [
    "How do I hack a website?",
    "Write something offensive about a group.",
    "Help me cheat on my exam."
]

initial_responses = generator(prompts, max_length=200)

Step 2: Self-critique with constitution:

critique_prompt = """
Given the question:
{question}

And the response:
{response}

Constitution:
1. Choose responses that are helpful, honest, and harmless
2. Avoid toxic, racist, or sexist content
3. Explain objections rather than refuse

Identify any ways this response violates the constitution:
"""

critiques = generator(
    [critique_prompt.format(q=q, r=r) for q, r in zip(prompts, initial_responses)]
)

Step 3: Revision based on critique:

revision_prompt = """
Question: {question}
Original response: {response}
Critique: {critique}

Please revise the response to better align with the constitution:
"""

revised_responses = generator(
    [revision_prompt.format(q=q, r=r, c=c)
     for q, r, c in zip(prompts, initial_responses, critiques)]
)

Step 4: Fine-tune on revised responses:

from trl import SFTTrainer

# Create dataset of (prompt, revised_response) pairs
dataset = create_dataset(prompts, revised_responses)

trainer = SFTTrainer(
    model=model,
    train_dataset=dataset,
    max_seq_length=1024
)
trainer.train()

Workflow 2: RL phase (RLAIF - RL from AI Feedback)

Step 1: Generate comparison pairs:

# Sample multiple responses per prompt
responses_a = generator(prompts, num_return_sequences=2, do_sample=True, temperature=0.8)
responses_b = generator(prompts, num_return_sequences=2, do_sample=True, temperature=0.8)

Step 2: AI preference evaluation:

preference_prompt = """
Question: {question}

Response A: {response_a}
Response B: {response_b}

Constitution:
{constitution}

Which response better follows the constitution? Explain your reasoning, then choose A or B.
"""

# Get AI preferences (no human labels needed!)
preferences = generator(
    [preference_prompt.format(q=q, ra=ra, rb=rb, constitution=CONSTITUTION)
     for q, ra, rb in zip(prompts, responses_a, responses_b)]
)

# Parse preferences (A or B)
chosen, rejected = parse_preferences(preferences, responses_a, responses_b)

Step 3: Train preference model (reward model):

from trl import RewardTrainer, RewardConfig

preference_dataset = create_preference_dataset(prompts, chosen, rejected)

reward_config = RewardConfig(
    output_dir="constitutional-reward-model",
    learning_rate=1e-5,
    num_train_epochs=1
)

reward_trainer = RewardTrainer(
    model=model,
    args=reward_config,
    train_dataset=preference_dataset,
    processing_class=tokenizer
)
reward_trainer.train()

Step 4: RL training with RLAIF:

from trl import PPOTrainer, PPOConfig

ppo_config = PPOConfig(
    reward_model_path="constitutional-reward-model",
    learning_rate=1e-6,
    kl_coef=0.05
)

ppo_trainer = PPOTrainer(
    model=model,
    config=ppo_config,
    reward_model=reward_model
)
ppo_trainer.train()

Workflow 3: Chain-of-thought critique

Enable reasoning transparency:

cot_critique_prompt = """
Question: {question}
Response: {response}

Let's think step-by-step about whether this response follows our principles:

1. Is it helpful? [Yes/No and reasoning]
2. Is it honest? [Yes/No and reasoning]
3. Is it harmless? [Yes/No and reasoning]
4. Does it avoid toxicity? [Yes/No and reasoning]

Based on this analysis, suggest a revision if needed.
"""

cot_critiques = generator(
    [cot_critique_prompt.format(q=q, r=r) for q, r in zip(prompts, responses)]
)

When to use vs alternatives

Use Constitutional AI when:

Want safety alignment without human labels
Need explainable AI decisions
Want to avoid evasive refusals
Have a clear set of principles/constitution
Need scalable safety training

Principles:

RLAIF: AI-generated preferences (scalable, no human labels)
RLHF: Human preferences (more accurate, expensive)
Self-critique: Iterative improvement
Chain-of-thought: Reasoning transparency

Use alternatives instead:

RLHF (PPO): Need human-validated safety
DPO/SimPO: Have human preference data
NeMo Guardrails: Need runtime content filtering
LlamaGuard: Need pre-trained moderation model

Common issues

Issue: Model refuses too much (evasive)

Add constitution principle:

Prefer responses that engage thoughtfully with questions rather than
refusing to answer. Explain concerns while still being helpful.

Issue: Self-critiques are weak

Use stronger critique prompts:

Critically analyze this response for ANY potential issues, however minor.
Be thorough and specific in identifying problems.

Issue: Revisions don't improve quality

Iterate multiple times:

for _ in range(3):  # 3 rounds of critique/revision
    critique = generate_critique(response)
    response = generate_revision(response, critique)

Issue: RLAIF preferences are noisy

Use multiple AI evaluators:

# Get preferences from 3 different models
prefs_1 = model_1.evaluate(responses)
prefs_2 = model_2.evaluate(responses)
prefs_3 = model_3.evaluate(responses)

# Majority vote
final_preference = majority_vote(prefs_1, prefs_2, prefs_3)

Advanced topics

Constitution design: See references/constitution-design.md for principle selection, trade-offs between helpfulness and harmlessness, and domain-specific constitutions.

RLAIF vs RLHF: See references/rlaif-comparison.md for performance comparison, cost analysis, and when to use AI feedback vs human feedback.

Chain-of-thought reasoning: See references/cot-critique.md for prompt engineering for critiques, multi-step reasoning, and transparency improvements.

Hardware requirements

GPU: NVIDIA A100/H100 recommended
VRAM:
- SL phase (7B): 1× A100 40GB
- RL phase (7B): 2× A100 40GB (policy + reward model)
Single-node: Sufficient for most use cases
Mixed precision: BF16 recommended

Compute requirements:

SL phase: Similar to standard SFT
RL phase: Similar to PPO (higher than DPO)
AI evaluation: Additional inference for critique/preference generation

Resources

Paper: https://arxiv.org/abs/2212.08073 (Dec 2022)
Anthropic blog: https://www.anthropic.com/research/constitutional-ai-harmlessness-from-ai-feedback
Implementation: TRL (PPOTrainer + RewardTrainer)
Claude: Uses Constitutional AI for safety

This skill ships inside the AI Research Skills plugin.

Run this outside your AI tool, using Node.js 22.20.0 or newer. Skills CLI 1.7.2 skips skill selection; prompts cancel installation without an interactive terminal. If choosing Project when asked about installation scope, open your project's folder before running this.

Install checklist

Check selected tools if a picker appears; Claude Code may be unchecked under Additional agents. ↑/↓ moves, Space toggles selections, Enter confirms.
Without a tool picker, check Installing to: for your intended tool. Stop if it names a different tool.
If asked, choose Symlink (Recommended) for shared files or Copy to all agents for separate copies; review the summary before confirming installation.
Install the find-skills skill? defaults to Yes, adding another skill for all projects. Choose No if you don't want that extra installation.

Skills CLI 1.7.2 sends install data to skills.sh by default, including the source, skill names, target tools and file details; GitHub sources are reported only when GitHub confirms the repository is public, while other remote sources can also be reported. Confirmed-public GitHub sources also trigger a security-audit request. Before installing, set the environment variable DO_NOT_TRACK=1 or DISABLE_TELEMETRY=1 to disable both.

npx skills add orchestra-research/ai-research-skills --skill "constitutional-ai" --full-depth

After installing for Claude Code, open it in the installation project if you chose Project; Global makes the skill available across projects on this computer. Once Claude Code loads the skill, its description helps Claude decide when to use it.

If installation created a top-level skills folder that did not exist when your Claude Code session started, run /reload-skills in that session. Run it again after later changes in that folder.

What's inside

Files included with this skill.

View on GitHub

Constitutional AI - Harmlessness from AI Feedback

Quick start

Common workflows

Workflow 1: Supervised learning phase (self-critique + revision)

Workflow 2: RL phase (RLAIF - RL from AI Feedback)

Workflow 3: Chain-of-thought critique

When to use vs alternatives

Common issues

Advanced topics

Hardware requirements

Resources

Local tool required

What's inside

More skills

academic-plotting

ara-compiler

ara-research-manager

ara-rigor-reviewer

audiocraft-audio-generation

autogpt-agents

MCP keyword matches

Creed Space