Anthropic has announced that its AI assistant Claude is now available to users in Brazil. The expansion brings the AI model to a new geographic market.
Claude, Anthropic’s trusted AI assistant, is now available in Brazil. Starting today, consumers and businesses in Brazil will be able to access Claude via:
Claude.ai offers both free and paid plans. Paid users on the Pro plan get 5x more usage versus our Free tier, early access to new features, and access to all our current Claude models—including Claude 3.5 Sonnet , our most intelligent model yet, and the rest of the Claude 3 family—for R$ 110 per user per month.
Paid users on the Team plan get even more usage and access to shared chats with teammates for R$ 165 per user per month, with a minimum of 5 seats. The Team plan includes administrative tools to conveniently control user and billing management for easier onboarding and less overhead.
Anthropic prioritizes safety, privacy, and ethical AI development in Claude's design. Rigorous testing and training aim to reduce misuse, while a strong commitment to user privacy means generative models are not trained on user data without explicit permission.
To evaluate the safety of our systems, Anthropic is giving employee-like access to a team from METR, a nonprofit that evaluates frontier AI models, and we're in early discussions with several other organizations, including Hugging Face.
Earlier this year, we shared an update on our election safeguards. Here, we outline how we’ve continued to strengthen these safeguards ahead of this year’s US midterms.
Expanding availability to new regions like Brazil allows AI companies to grow their user base and compete in international markets.