L0 · LabsFrontier labUnited States

Anthropic

Safety-focused lab; maker of Claude

Anthropic builds the Claude family with a research emphasis on interpretability, Constitutional AI, and safety, and is the leader in enterprise/coding usage via Claude Code and API. It runs multi-cloud on AWS (primary, using Trainium) and Google Cloud (using TPUs). In 2025-2026 it raised large rounds led by strategic backing from Amazon (~$8B total) and Google (~$3B+), and shipped Claude Opus and Sonnet generations targeting agentic and coding workloads.

Flagship model

Claude (Opus / Sonnet)

Primary cloud

AWS (Trainium) + Google Cloud (TPU)

Lead investors

Amazon, Google

Strength

Enterprise, coding, safety

How it fits the stack

Anthropic with what it depends on (above) and what it feeds (below). The figure renders as a crawlable diagram and upgrades to an interactive 3D graph as it scrolls into view.

AnthropicDepends on ↑Feeds ↓Related

Anthropic in the AI stack. Anthropic with its immediate upstream dependencies (top) and downstream dependents (bottom) in the AI value chain. Hover a node in 3D, or read the full relationships below.

Graph data (text) — 16 entities, 18 relationships

Anthropic —depends on→ Amazon Web Services (AWS)
Anthropic —hosts→ Amazon Web Services (AWS)
Anthropic —uses→ AWS Trainium / Inferentia
Anthropic —used by→ AWS Trainium / Inferentia
Anthropic —depends on→ Google Cloud (GCP)
Anthropic —hosts→ Google Cloud (GCP)
Anthropic —used by→ Google TPU
AI coding agents & IDEs —depends on→ Anthropic
AI Coding Assistants (Cursor, Copilot, Windsurf, Claude Code) —supplies→ Anthropic
Cognition (Devin) —depends on→ Anthropic
Cursor (Anysphere) —depends on→ Anthropic
GitHub Copilot —uses→ Anthropic
Glean —uses→ Anthropic
Harvey AI —uses→ Anthropic
Anthropic —partners with→ AI Safety & Interpretability Research (ARC, Redwood, Apollo)
Anthropic —competes with→ Amazon AGI (Nova)
Anthropic —partners with→ Anduril / Defense AI
Anthropic —partners with→ Defense & government AI

Depends on ↑ · 22

depends onAmazon Web Services (AWS)AWS is Anthropic's primary cloud; trains on Trainium hostsAmazon Web Services (AWS)Primary cloud + Project Rainier Trainium2 cluster for Anthropic training usesAWS Trainium / InferentiaProject Rainier Trainium2 cluster for Claude training used byAWS Trainium / InferentiaAnthropic trains on Trainium2 (Project Rainier)depends onGoogle Cloud (GCP)Anthropic also uses Google Cloud TPUs at large scale hostsGoogle Cloud (GCP)2025 deal for up to ~1M TPUs / ~1GW+ capacity for Anthropic used byGoogle TPUAnthropic committed to large TPU capacity suppliesMercorexpert-domain labelers for post-training/evals suppliesMLOps / experiment & observabilityExperiment tracking / observability depends onNvidia Data-Center GPU (Blackwell/Rubin)⬦Also uses Nvidia GPUs alongside Trainium/TPU used byNvidia Data-Center GPU (Blackwell/Rubin)⬦Anthropic uses Nvidia GPUs alongside TPU/Trainium usesPyTorchPyTorch/JAX are the substrate above the accelerator layer depends onScale AI⬦RLHF & evaluation data suppliesScale AI⬦Expert labeling and evals sourcesSurge AIRLHF/human-preference data from neutral vendor after Meta absorbed Scale suppliesSurge AIPremium RLHF/human preference data to frontier labs depends onSurge AI / Mercor (expert data)PhD-level domain annotation suppliesSurge AI / Mercor (expert data)Expert human data / RLHF supplier to labs usesSurge AI / Mercor / Turing (frontier data labor)Expert human SFT/RLHF contractors used bySWE-bench / Evaluation Benchmarkscoding benchmark used to evaluate Claude usesSynthetic data & distillationConstitutional AI and distillation use synthetic data suppliesTraining Data & Web Corpus⬦data access/provenance is a contested non-hardware chokepoint

Feeds ↓ · 9

depends onAI coding agents & IDEsTop revenue vertical shaping model roadmap suppliesAI Coding Assistants (Cursor, Copilot, Windsurf, Claude Code)Coding is a top Anthropic revenue channel depends onCognition (Devin)Devin relies heavily on Claude for coding depends onCursor (Anysphere)large share of Anthropic API revenue; token demand driver usesGitHub CopilotMulti-model: offers Claude to developers usesGleanEnterprise RAG assistant on frontier models usesHarvey AIMulti-model legal deployments usesLegal AIHarvey/CoCounsel built on frontier models usesPublic Sector & Government AIGSA OneGov $1 Claude access

Context — capital, rivals, policy · · 30

partners withAI Safety & Interpretability Research (ARC, Redwood, Apollo)Interpretability/scheming research informs deployment competes withAmazon AGI (Nova)Amazon's first-party Nova competes with its own investee on Bedrock partners withAnduril / Defense AIClaude for defense via Palantir/AWS gov deals partners withDefense & government AICDAO / DoD frontier contracts partners withDefense & National Security AI⬦Claude in classified/gov environments partners withEnterprise Data Cloud (Snowflake/Databricks)⬦Snowflake Cortex / Databricks model access competes withGoogle DeepMindCompete for enterprise despite Google's investment partners withMETRAutonomy evals with pre-release access competes withOpenAIDirect frontier and enterprise/coding rivals partners withPalantir AIPAIP deploys Claude in enterprise/gov via ontology partners withPalantir TechnologiesAIP + Claude for defense/enterprise deployment on AWS partners withSalesforce AgentforceAgentforce integrates Anthropic models partners withUK AI Safety Institute (AISI)Pre-deployment model evaluation access invests inAmazon$8B invested; primary compute partner invests inAmazon AGI (Nova)Amazon invested ~$8B in Anthropic; primary cloud partner invests inAmazon Web Services (AWS)~$8B+ total strategic investment in Anthropic invests inAndreessen Horowitz (a16z)holds Anthropic exposure among model-layer bets invests inAnthropic capital stack (Amazon/Google)~$183B-round capital funds Claude training compute invests inGoogle / Alphabet$3B+, ~14% stake, while competing via Gemini invests inGoogle Cloud (GCP)Multiple equity investments in Anthropic invests inGoogle DeepMindGoogle invested ~$3B+ in Anthropic and provides TPUs restrictsAI Copyright Litigation (NYT et al.)⬦~$1.5B Anthropic books settlement (2025)restrictsAI Safety Institutes (US CAISI / UK AISI)Pre-deployment frontier evals restrictsCopyright & AI litigation⬦Bartz v. Anthropic settlement shapes data use restrictsEU AI Acttransparency/risk duties apply to frontier providers in EU restrictsEU AI Act GPAI Code of PracticeSigned the GPAI Code of Practice restrictsUS AI Safety Institute (CAISI/NIST)Pre-deployment evaluation of frontier models designsClaude (Opus / Sonnet)Anthropic designs the Claude Opus/Sonnet/Haiku family designsFrontier Safety Frameworks (RSP / Preparedness / FSF)Responsible Scaling Policy / ASL tiers designsModel Context Protocol (MCP)Anthropic authored MCP; adopted by OpenAI/Google/Microsoft

In the news · 2

Anthropic · 2026-05-19

Andrej Karpathy Joins Anthropic's Pre-Training Team

The OpenAI founding member and former Tesla AI director has joined Anthropic to work on pre-training, building a group that uses Claude to accelerate pretraining research. Eureka Labs, the education company he founded in 2024, is paused rather than closed.

Anthropic · 2026-07-20

Court Gives Final Approval to Anthropic's $1.5 Billion Author Settlement

A federal judge granted final approval to Anthropic's $1.5 billion settlement with book authors and publishers on July 20, 2026, the largest known copyright settlement in U.S. history. Roughly 500,000 pirated books, about $3,000 per work, and a fair use ruling on AI training that survives intact.

← Back to the AI Systems Map