Claude
Anthropic · United States
What it is
Anthropic's general-purpose AI assistant for writing, analysis, drafting, research and coding. You reach it through the Claude apps (web, desktop and mobile), where it offers chat, Projects (a workspace that keeps your documents and instructions to hand), Artifacts (editable documents and code shown beside the conversation), web search, file and image analysis, and connectors to outside tools. The paid plans add Cowork, an agent mode that carries out whole tasks for you: it can work with files in folders you choose to connect, use your connectors and, since August 2026, operate a browser of its own. It also comes as Claude Code, a tool for software development, and as an API that many other products are built on.
How it works
Claude is a product built on Anthropic's Claude family of large language models, which now comes in four sizes: Fable (the frontier tier, sitting above Opus), Opus (the most capable of the established models), Sonnet (the everyday balance) and Haiku (the fastest). (The version numbers move quickly, so we describe the family rather than a release that will date.) Around the model sit the product features: Projects, web search, file upload, memory, connectors and Cowork, the agent mode that since August 2026 includes a browser Claude operates itself. How your data is handled mostly depends on your plan rather than the model. Fable is the one exception: it carries a mandatory 30-day retention that the established models do not (see our alerts).
Who is behind it
- Owner
- Anthropic, a United States artificial-intelligence company structured as a public-benefit corporation (Anthropic PBC). It was founded by a group of former OpenAI staff, and its largest investors are Amazon and Google. CEO Dario Amodei; President Daniela Amodei.
- Founded
- Founded in 2021 by Dario and Daniela Amodei and other former members of OpenAI; the first Claude models were released in 2023.
- Headquarters
- San Francisco, California, United States; the consumer terms are governed by California law.
Plans
- Claude Free / Pro / Max · Consumer (free / paid)Personal accounts; a Privacy Settings toggle decides whether new chats train models (off: no training, 30-day retention; on: training, retention up to five years).
- Claude Team · TeamSmall-business plan under the Commercial Terms; no training on inputs, but no custom retention control or audit log.
- Claude Enterprise · Enterprise / WorkspaceNo training on inputs; adds retention controls, audit logs and a HIPAA-ready (BAA) option.
- Anthropic API (incl. zero data retention) · API directThe model layer for developers; no training on inputs, zero data retention available.
How it handles your data
The plan is what matters. The consumer plans — Free, Pro and Max — are treated the same for data purposes, and the control that counts is a model-training toggle in Privacy Settings, available even on the free plan: turn it off and your new chats are not used to train models and retention drops to 30 days; leave it on and Anthropic may use them to train models, with retention of up to five years. Either way a consumer account is a personal one — there is no data-processing agreement, and conversations flagged by Anthropic's safety systems may be reviewed by staff — and paying for Pro or Max buys more usage, not different data protections. The commercial plans — Team, Enterprise and the API — run under Anthropic's Commercial Terms, which do not train on your inputs. Team adds a shared workspace and admin tools but no custom retention control or audit log; Enterprise adds retention controls, audit logs and a HIPAA-ready option (with a BAA) for health data; and the API does not train on inputs and can be configured for zero data retention (except on Fable 5, whose mandatory 30-day retention applies on every platform and cannot be contracted out of). Anthropic processes in the United States and does not offer a New Zealand or Australian data region, so putting personal information in remains a cross-border disclosure risk (see the grid below).
Sources for this description
- Consumer Terms of Service (effective 8 October 2025) — Anthropiclive sourcePerma.cc recordas at 1 June 2026
- Updates to Consumer Terms and Privacy Policy — Anthropiclive sourcePerma.cc recordas at 30 May 2026
- Commercial Terms of Service (effective 17 June 2025) — Anthropiclive sourcePerma.cc recordas at 30 May 2026
- How Do You Use Personal Data in Model Training? — Anthropiclive sourcePerma.cc recordas at 30 May 2026
- Configure custom data retention controls for Enterprise plans — Anthropiclive sourcePerma.cc recordas at 30 May 2026
- How do I change my model improvement privacy settings? — Anthropiclive sourcePerma.cc recordas at 30 May 2026
- Data retention practices for Mythos-class models — Anthropiclive sourcePerma.cc recordas at 11 June 2026
- Claude Cowork gets a built-in browser (announcement, 26 August 2026) — Anthropiclive sourceas at 28 August 2026
What is safe, on which plan
Each cell summarises what the provider’s terms and the applicable rules say. The colour is our summary of the sources, not advice — select a cell to read the reasoning and follow the sources yourself.
- Green
- Amber
- Red
- Black
- Not yet assessed
| Data category | Consumer (free / paid) | Team | Enterprise / Workspace | API direct |
|---|---|---|---|---|
| 1. Public information | ||||
| 2. Internal non-confidential | ||||
| 3. Internal confidential (non-privileged) | ||||
| 4. Legally privileged | ||||
| 5. Personal information (Privacy Act 2020) | ||||
| 6. Sensitive personal information | ||||
| 7. Children’s information | ||||
| 8. Court-protected material |
1. Public information
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
2. Internal non-confidential
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
3. Internal confidential (non-privileged)
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
4. Legally privileged
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
5. Personal information (Privacy Act 2020)
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
6. Sensitive personal information
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
7. Children’s information
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
8. Court-protected material
- Consumer (free / paid)
- Team
- Enterprise / Workspace
- API direct
Select any cell to see what the sources say and why.
Statements about provider terms, plans, and defaults will go stale. This profile names its last review date above.
Where to next
Our framework
The reference behind every answer: the eight kinds of information, the five plans, and what each colour means.
Our frameworkGuidance
Our view on the six questions to ask before you use a tool — confidentiality, privilege, privacy, connected apps, human review, and the professional rules.
GuidanceAbout us
Who is behind this, how we keep it reliable, and why you can trust it.
About us