**Claude** is a family of [[Large language model|large language models]] developed by [[Anthropic]], an American [[Artificial intelligence|AI]] safety company. First released in March 2023, Claude is designed to be helpful, harmless, and honest, reflecting Anthropic's research focus on developing safe and beneficial AI systems. Claude is trained using a technique called [[Reinforcement learning from human feedback|reinforcement learning from human feedback]] (RLHF) combined with Anthropic's proprietary method known as Constitutional AI (CAI), which aims to make AI systems more controllable and aligned with human values. The model is available through Anthropic's consumer-facing chatbot interface, Claude.ai, as well as through an [[Application programming interface|API]] for developers. As of 2024, the Claude model family includes multiple versions optimized for different use cases, including Claude Opus, Claude Sonnet, and Claude Haiku, which vary in capability and computational cost. Anthropic positions Claude as a competitor to other prominent AI assistants, including [[OpenAI]]'s [[ChatGPT|ChatGPT]] and [[Google]]'s [[Gemini (chatbot)|Gemini]]. The company has raised significant funding from investors including [[Google]] and [[Salesforce]], and Claude has been integrated into various enterprise and consumer applications. Claude's development reflects broader industry trends toward creating AI systems that can engage in extended reasoning, follow complex instructions, and assist with tasks ranging from writing and analysis to [[Computer programming|coding]] and research.