Constitutional AI: how Anthropic builds safety in

Claude is trained using Constitutional AI (CAI), a technique Anthropic invented that supplements human preference labeling with a written 'constitution' — a set of principles the model uses to critique and revise its own outputs. This lets Anthropic scale alignment without requiring humans to hand-label millions of harmful examples, and it makes Claude's refusals more principled: it can usually explain why it will not do something rather than refusing blankly.

• Step 1 (SL-CAI): the model answers red-team prompts, then critiques and rewrites its own responses against the constitution
• Step 2 (RL-CAI): a preference model is trained on those AI-generated critiques and used as the reward signal for reinforcement learning
• The constitution draws on sources such as the UN Declaration of Human Rights plus Anthropic's own principles
• Result: more consistent, explainable alignment that scales without armies of human labelers

CAI vs standard RLHF: Standard RLHF relies entirely on human raters to mark what is harmful. Constitutional AI adds model self-critique against explicit principles, which makes alignment more consistent and far easier to scale.

The Claude model family: from Claude 3 to the 4.x generation

Anthropic uses three named tiers within each generation — Haiku (fast and cheap), Sonnet (balanced default), and Opus (maximum capability) — plus, as of June 2026, a Mythos-class tier above Opus. A consistent pattern holds across generations: each new Sonnet tends to match or beat the previous generation's Opus at a lower price.

Where the line is now (mid-2026): Claude 3.5 Sonnet is historic, not current. Today's pricing is Opus 4.8 at $5 / $25, Sonnet 4.6 at $3 / $15, and Haiku 4.5 at $1 / $5 per million input/output tokens, all with a 1M-token context window on the 4.6+ models. Fable 5 sits above Opus but had access restricted in mid-2026 under a US export-control directive.

Extended and adaptive thinking

Claude 3.7 Sonnet introduced 'extended thinking' — the model reasons step by step inside a separate token budget before producing its visible answer, similar in spirit to OpenAI's o-series. In the 4.x generation this evolved into adaptive thinking: instead of toggling a mode, you ask for harder reasoning and the model scales its own depth, spending more internal compute only when the task warrants it.

import anthropic

client = anthropic.Anthropic()

response = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=16000,
    thinking={"type": "enabled", "budget_tokens": 10000},
    messages=[{"role": "user",
               "content": "Prove that the square root of 2 is irrational."}],
)

for block in response.content:
    if block.type == "thinking":
        print("REASONING:", block.thinking[:200])
    elif block.type == "text":
        print("ANSWER:", block.text)

What Claude is best at

• Agentic coding and large-codebase work — Claude Code is one of the most widely used AI coding tools
• Long-document analysis using the 1M-token context window
• Careful instruction following and brand-safe, nuanced writing
• Safety-sensitive deployments where consistent, explainable refusals matter

For most production work Sonnet 4.6 is the cost-effective default; Opus 4.8 is reserved for the hardest reasoning and coding; Haiku 4.5 handles high-volume, latency-sensitive traffic. A routing strategy across all three usually beats defaulting everything to the most capable model.

Benchmarks and key specs

How fast this moved: Claude 3.5 Sonnet set a coding-benchmark record at launch in 2024 with about 49% on SWE-bench Verified. Less than two years later, Sonnet 4.6 reaches ~79.6% and Opus 4.8 reaches 88.6% — a vivid illustration of how quickly the frontier moved. Numbers are vendor-reported.

Frequently asked questions

Is Claude 3.5 Sonnet still the latest? No. It was a milestone in 2024, but the current family centers on Claude Opus 4.8, Sonnet 4.6, and Haiku 4.5, with a Mythos-class tier (Fable 5) above Opus.

What is Constitutional AI? It is Anthropic's alignment method: the model critiques and revises its own outputs against a written set of principles (a 'constitution'), which scales alignment without relying solely on human labels and makes refusals more principled.

Which Claude model should I use? Sonnet 4.6 is the cost-effective default for most coding and analysis; Opus 4.8 is for the hardest reasoning and large-codebase work; Haiku 4.5 handles high-volume, latency-sensitive tasks.

Does Claude have a free tier? Yes — claude.ai offers free access with limits, while paid Pro and Max plans raise usage caps and unlock the most capable models.