Anthropic

Profil

Anthropic adalah perusahaan AI asal San Francisco, California yang fokus pada AI safety. Pembuat model Claude dan pemimpin agentic AI 2025-2026.

Sejarah

• 2021 — didirikan oleh Dario Amodei & Daniela Amodei, mantan VP Research & VP People di OpenAI
• 2022 — Claude 1 (private beta)
• 2023 — Claude 1 (public), Claude 2
• 2024 — Claude 3 (Haiku, Sonnet, Opus), Claude 3.5 Sonnet
• 2024 — Computer Use (agentic AI pioneer)
• 2025 — Claude 4, 4.5 Sonnet, Claude Code
• 2025 — valuation $200B+, ARR $5B+
• 2026 — Frontier agents, mechanistic interpretability breakthroughs

Produk

• Claude (chatbot & API)
• Claude Code (agentic coding di terminal)
• Computer Use (browser & OS control)
• MCP (Model Context Protocol — standar terbuka)
• API

Pendanaan

• 2022 — $580M Series B
• 2023 — $1.25B (Google, Salesforce, Amazon)
• 2024 — $4B (Amazon $4B, total $8B Amazon)
• 2025 — $2B (Lightspeed, Bain)
• Valuation: $60B (awal 2025), $200B+ (akhir 2025)

Konsep Kunci

• Constitutional AI — alignment dengan prinsip
• Mech-Interp — mechanistic interpretability
• Responsible Scaling Policy — safety framework
• AI welfare — apakah AI consciousness harus dipertimbangkan

Klien Enterprise

Bank of America, Pfizer, Deloitte, GitLab, Slack, Zoom, Bridgewater, dll.