Claude
Anthropic released its own conversational model on the same day as GPT-4, betting on steerability and less harmful output.
Why it matters
The frontier model market stopped being a monopoly in a single day, and the rival came from people who had left OpenAI.
Anthropic was founded in 2021 by former OpenAI staff, among them Dario and Daniela Amodei. Its alignment approach is called constitutional: the model critiques its own answers against a written set of principles instead of relying on human raters alone. The coincidence of dates was not accidental and set the rhythm of the following years, when competitors' announcements came back to back.