Anthropic
Category: Foundation Models / LLMs
Anthropic builds Claude, a family of frontier large language models designed with safety at the core, offering reliable, interpretable, and steerable AI systems for enterprises, developers, and consumers. Anthropic was founded in 2021. The company is led by Dario Amodei. Based in San Francisco, United States. Team size: 2,500-4,000. Total funding raised: $132.3B. Latest round: Series G. Key investors include GIC, Coatue, ICONIQ Capital, Altimeter Capital, Dragoneer, Greenoaks, Sequoia Capital, Capital Group, D1 Capital Partners, XN, Amazon, Google, Microsoft, NVIDIA, Lightspeed Venture Partners, Fidelity Management & Research, D. E. Shaw Ventures, Founders Fund, Menlo Ventures, Salesforce Ventures, MGX, Temasek, Blackstone, Brookfield, Insight Partners, General Catalyst, Jane Street, Baillie Gifford, DST Global, AMP PBC, T. Rowe Price, Micron, Samsung, SK hynix.
- Founded
- 2021
- Headquarters
- San Francisco, United States
- Team size
- 2,500-4,000
- Total funding
- $132.3B
Value proposition
Combines frontier AI capabilities with Constitutional AI methodology to deliver models that are helpful, harmless, and honest—enabling enterprises to deploy AI confidently in regulated and high-stakes environments.
Products and solutions
Claude Model Family (Opus 4.8, Sonnet 4.8, Haiku 4.5, Mythos Preview), Claude API Platform, Claude Code (AI coding agent), Claude Cowork (knowledge work agent), Claude Enterprise & Claude Team subscriptions, Claude Marketplace (third-party software ecosystem), Claude for Excel integration, Model Context Protocol (MCP) extensions, Dynamic Workflows (subagent orchestration)
Unique value
Pioneered Constitutional AI methodology—training models through self-improvement guided by explicit principles (a 'constitution') rather than relying solely on human feedback. Published Claude's constitution publicly in January 2026 for transparency.
Target customer
Enterprise organizations, software developers, government agencies, and individual users requiring safe, high-performance AI for mission-critical applications
Industries served
Financial Services & Banking, Healthcare & Life Sciences, Government & Defense, Technology & Software Development, Legal & Professional Services, Manufacturing, Retail & E-commerce, Education
Technology advantage
Combines frontier model performance with built-in safety alignment through Constitutional AI; offers extended thinking mode for complex reasoning, computer use capabilities for autonomous task execution, and multi-modal processing. Claude Opus 4.6 achieves industry-leading coding and agentic capabilities while maintaining safety guardrails.
How they differentiate
Pioneered Constitutional AI methodology training models through self-improvement guided by explicit principles rather than solely human feedback; Public Benefit Corporation structured for societal good; Safety-first 'Helpful, Harmless, and Honest' approach with transparent AI constitution; Enterprise-first go-to-market strategy versus consumer-first competitors
Main competitors
OpenAI, Google DeepMind, Meta AI
Key partnerships
Amazon ($5B+ investment, AWS Bedrock, up to 5GW compute capacity), Google ($2.5B investment, Google Cloud Vertex AI, Broadcom TPU deal for 5GW), Microsoft (Azure Foundry integration), Broadcom (5GW next-gen TPU capacity partnership, announced April 2026), SpaceX (GPU capacity in Colossus 1 and Colossus 2), Accenture (multi-year partnership, largest Claude Code deployment), Snowflake ($200M multi-year agreement), Salesforce (Agentforce integration for regulated industries), Deloitte (enterprise AI deployment alliance), Palantir & AWS (U.S. government/defense sector), Micron, Samsung, SK hynix (strategic infrastructure partners)
Notable customers
8 of Fortune 10 (undisclosed), Accenture (largest Claude Code deployment), Snowflake ($200M multi-year agreement), Salesforce (Agentforce integration), Commonwealth Bank of Australia, Palantir (government/defense), Deloitte, Cognizant, Intercom, NASA (Claude on Mars - Perseverance rover, Jan 2026)
Major milestones
Founded 2021 by former OpenAI researchers including Dario and Daniela Amodei, Pioneered Constitutional AI methodology and published Claude's constitution publicly (January 2026), Series A raised $124M at $550M valuation (May 2021), Amazon strategic partnership with $8B total investment (2024), Launched Claude 3 model family achieving industry-leading benchmarks, Series E raised $3.5B at $61.5B valuation led by Lightspeed (March 2025), Series F raised $13B at $183B valuation led by ICONIQ (September 2025), Acquired Coefficient Bio for $400M - first acquisition entering biotech AI (April 2026), Series G raised $30B at $380B valuation - second-largest venture deal ever (February 2026), Revenue run rate crossed $30B (April 2026), then $47B (May 2026), Launched Claude Mythos Preview - frontier model for cybersecurity (April 2026), Series H raised $65B at $965B valuation - surpassed OpenAI as most valuable AI startup (May 2026), Launched Claude Opus 4.8 with Dynamic Workflows for subagent orchestration (May 2026), 8 of Fortune 10 now Claude customers; 500+ customers spending $1M+ annually
Growth metrics
$47B+ annual run rate (May 2026); $30B run rate crossed in April 2026; 80x year-over-year growth in Q1 2026; $965B valuation post-Series H (May 2026); 8 of Fortune 10 are Claude customers; 500+ customers spending $1M+ annually; Claude Code run-rate revenue over $2.5B; 4% of all GitHub public commits authored by Claude Code
Market positioning
Enterprise AI market leader with 32%+ share of enterprise LLM spending, surpassing OpenAI; Positioned as trusted intelligence platform for mission-critical regulated industries (financial services, healthcare, government); $965B valuation post-Series H (May 2026), surpassing OpenAI as most valuable AI startup; 8 of Fortune 10 as customers; Revenue run rate $47B+
Geographic focus
Global enterprise presence with headquarters in San Francisco, expanding operations across North America, Europe, and Asia-Pacific; 300,000+ business customers worldwide with localized compliance features for different regulatory regimes
Patents and IP
61 patents across eight jurisdictions (primarily defensive portfolio, mostly acquired); focus on cloud services and AI infrastructure. Also reached $1.5B copyright settlement with authors in September 2025, establishing precedent for licensed training data.
About Dario Amodei
VP of Research at OpenAI (2016-2020) where he led teams responsible for GPT-2 and GPT-3 development; Senior Research Scientist at Google Brain (2015-2016); Researcher at Baidu; PhD in Physics from Princeton University (2006-2011); Postdoctoral Fellow at Stanford University School of Medicine (2011-2014); BS in Physics from California Institute of Technology (2001-2005)
Official website: https://www.anthropic.com