Anthropic Releases Claude — A Safety-First Rival LLM

On 14 March 2023, the San Francisco AI startup Anthropic released its conversational large language model Claude to the public.

On the same day, OpenAI announced GPT-4.

The coincidence of two leading models shipping on one date was a fair preview of the structure that would follow. The capital and the talent that believed generative AI needs a counter-pole would gather around these two companies.

Departure from OpenAI

Anthropic's core had, until early 2021, been the core of OpenAI.

• Dario Amodei: OpenAI's VP of Research and a lead author of the GPT-3 paper.
• Daniela Amodei: OpenAI's VP of Safety and Policy. Dario's sister.
• Tom Brown: the GPT-3 team lead.
• Sam McCandlish, Jack Clark, Jared Kaplan: researchers on GPT-3 and on the scaling laws.

In late 2020 and early 2021 they left OpenAI as a group. The formal reason was never stated; the contour that emerged from reporting was that after the 2019 Microsoft exclusivity deal, OpenAI was prioritising commercialisation and speed at the expense of safety.

Anthropic was incorporated in May 2021. Seed funding was US$124 million. Headcount at founding: seven.

Constitutional AI

What set Anthropic apart was less a particular technical innovation than the explicit articulation of a safety method.

In a December 2022 paper, Constitutional AI: Harmlessness from AI Feedback, Anthropic introduced a training approach. Simplified:

1. The AI generates a response.
2. Another AI critiques that response against a "constitution"—a set of ethical principles.
3. Based on the critique, the AI revises its response.
4. The loop converges on safer responses without continuous human labelling.

Unlike RLHF (OpenAI's mainline method), which routes through human raters, Constitutional AI has the model internalise the constitution. It both reduces the cost of human evaluation and makes "why this response was chosen" explicitly inspectable.

Claude was the first major dialogue model trained with Constitutional AI.

A Chain of Investments

What was unusual about Anthropic was that, while positioning itself as a safety-led research lab, it attracted the second-largest investment in generative AI.

• May 2023: Google invests US$400 million.
• September 2023: Amazon announces up to US$4 billion, of which US$1.25 billion within the year.
• October 2023: Google adds US$2 billion.
• 2025: Google invests further (cumulative reportedly over US$6 billion).

The logic for Amazon and Google was simple: against the Microsoft–OpenAI alliance, supporting Anthropic was the only credible option.

The Claude family expanded to Claude 3 (Haiku / Sonnet / Opus) in March 2024, Claude 3.5 Sonnet in June 2024, Claude 3.7 Sonnet with an extended-thinking mode in February 2025, and the Claude 4 series by May 2025. In coding and agentic benchmarks, Claude began outperforming OpenAI's models.

What "Another Pole" Meant

The shared 14 March 2023 launch date was a coincidence, but a telling one. The OpenAI–Anthropic duality would shape the terrain of the AI industry:

• OpenAI: speed and scale, deep partnership with Microsoft, commercial maximisation.
• Anthropic: safety and explainability, distributed partnerships with Google and Amazon, expansion into government and defence contracts (the November 2024 Palantir–AWS coalition).

The two were competitors. But for the industry as a whole, they also divided the role of preventing single-vendor dominance. From the autumn of 2023, when regulatory debate intensified, the fact that governments and researchers had two contrasting frontiers to point to owes a great deal to Anthropic's existence.

And the founders of both companies—who once shared the same OpenAI offices, building GPT-3—now build competing frontier models from separate buildings.