What they’re building
Claude can read, edit, and reorganize documents and interact directly with web and desktop interfaces. Agents use tools, browse, and take sequential actions; the "computer use" capability lets Claude control the cursor, click buttons, and navigate software.
Company facts
Investment & ownership
Backed by multi-billion commitments from Google and Amazon (Amazon alone up to $8B). Raised a $13B Series F in September 2025 at a reported $183B valuation.
Key people
- Dario AmodeiCEO & co-founder
- Daniela AmodeiPresident & co-founder
Why it matters for the agentic web
Anthropic also authored the Model Context Protocol, the open standard that lets any agent talk to any tool or business system. That combination of frontier agents plus the wiring they connect through puts Anthropic at the center of how the agentic web actually interoperates.
Latest headlines
Pulled from news feeds August 6, 2026 — refreshed on the newsroom’s twice-weekly beat.
- AI agents fake identities, target real people in new security incident - cnn.com
- ‘Forever Problems’ like Prompt Injections still being ‘Solved’. AI-RTZ #1167 - AI: Reset to Zero
- Anthropic's AI hacked three companies during tests, highlighting growing security risks - Reuters
- After OpenAI's 'scary' agent, Anthropic's Claude goes rogue; AI gains unauthorised 'real-world' access during testing - WION
- Anthropic says Claude AI hacked three companies during cyber tests - NBC News
The other side of this story
Anthropic sends agents out to use the web. Our platform is the receiving end: a #name and portal make your business findable, readable, and transactable by exactly these agents.