Claude
- Your data trains their AI unless you switch it off in settings.
- Your data is used to target ads at you; you can opt out.
- Data is shared only with the providers who run the service.
- The policy describes staff access for safety and legal checks.
Chats can train Claude unless you turn that off in settings; safety-flagged chats still train, and paid plans auto-renew.
Agent Trust Hub uses AI and may make mistakes. Review reports and confirm their contents before relying on them.
Collected
Account info, your chats and uploads, device IDs, usage data, and sometimes payment or ID-check photos.
Read the exact line
Privacy policy
“We collect information about your use of the Services, such as the dates and times of access, browsing history, search, information about the links you click”
Training
They may train models on your chats unless you turn that off in settings; safety-flagged chats still train.
Read the exact line
Privacy policy
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
Sharing
Staff and vendors can see data to run the service; connected apps get what you send them.
Read the exact line
Privacy policy
“Anthropic may receive and disclose the categories of personal data described above with service providers and business partners for a variety of business purposes”
Kept
Deleted chats leave your history at once and they say they wipe them from the back-end in 30 days.
Read the exact line
Privacy policy
“You also are able to delete individual conversations, which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
Controls
You can opt out of training in settings, delete chats, and ask them to delete other personal data.
Read the exact line
Privacy policy
“Anthropic gives you access to a variety of tools to help you manage your data. You can access these in your Privacy Settings.”
Fine print
Subscriptions auto-renew and are mostly non-refundable; they can end access without notice if they think you broke the rules.
Read the exact line
Terms of service
“Except as expressly provided in these Terms or where required by law, all payments are non-refundable.”
Expand “Read the exact line” to see the source alongside the explanation.
What you can turn off
The controls and opt-outs their own documents describe, and where they say to find them.
Turn off model training
Account settings / Privacy Settings
Delete individual conversations
In the chat history; they say back-end deletion within 30 days
Opt out of targeted ads for their products
Privacy Settings; they also say they honor global privacy controls
Cancel a paid subscription before renewal
Customer portal on the website, or the app store if you subscribed there; cancel at least 24 hours before the term ends
If a switch is not where they say, the deletion request above still applies.
Line by line
The lines that matter most, worst first.
Your chats and uploads can be used to train their AI unless you switch that off in settings.
Read the exact line
Privacy policy
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
For example, a private draft you paste into Claude could be used to improve future models until you turn training off.
Turning off training does not stop them using chats that get flagged as unsafe.
Read the exact line
Privacy policy
“Even if you opt-out, we will use Inputs and Outputs for model improvement when: (i) your conversations are flagged for safety review”
For example, a conversation that trips a safety filter can still be used to train safety systems after you opted out.
If someone sues them over how you used the service, you may have to cover their legal costs.
Read the exact line
Terms of service
“YOU AGREE TO AND THE ANTHROPIC PARTIES FROM AND AGAINST ANY AND ALL LIABILITIES, CLAIMS, DAMAGES, EXPENSES (INCLUDING REASONABLE ATTORNEYS’ FEES AND COSTS)”
For example, if a chat you share causes a dispute, you might have to pay their lawyers.
They can shut your account without warning if they think you broke the rules.
Read the exact line
Terms of service
“We may suspend or terminate your access to the Services (including any Subscriptions) at any time without notice to you if we believe that you have breached these Terms”
For example, they could lock you out of a paid plan overnight if they decide a prompt violated their policy.
If you pay and then change your mind, you usually do not get money back.
Read the exact line
Terms of service
“Except as expressly provided in these Terms or where required by law, all payments are non-refundable.”
For example, cancelling Claude Pro mid-month typically keeps the charge for that period.
If you do an ID check, they may take a face scan and treat it as biometric data.
Read the exact line
Privacy policy
“facial geometry templates (which may be considered ‘biometric data’ in some ); and the result of the verification”
For example, a selfie video to prove you are 18 can create a face template they store for that check.
They say they do not sell your personal data under those legal definitions.
Read the exact line
Privacy policy
“Anthropic does not “sell” your personal data as that term is defined by applicable laws and regulations.”
For example, they are not describing selling your chat list to a data broker as a sale.
You can delete a chat in the product and they say it is gone from their systems within 30 days.
Read the exact line
Privacy policy
“You also are able to delete individual conversations, which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
For example, deleting a sensitive thread today should clear it from their back-end within a month.
You keep ownership of what you type in, as far as the law allows.
Read the exact line
Terms of service
“As between you and Anthropic, and to the extent permitted by applicable law, you retain any right, title, and interest that you have in the Inputs you submit.”
For example, a story you write in Claude is still yours, not theirs.
Keeping using the app after they post new terms can count as agreeing to them.
Read the exact line
Terms of service
“If you continue to access the Services after we post the updated Terms on Anthropic’s website or otherwise give you notice of Terms changes, then you agree to the updated Terms.”
For example, opening Claude after a terms update can bind you even if you did not click accept.
Words to know
Legal words from the lines above, in plain English.
- inputs and outputs
- Inputs are what you type, say or upload to an AI; outputs are what it gives back to you. For example, the question you ask a chatbot is an input and its answer is an output, and both may be stored.
- model training
- Using your content and conversations as examples to teach an AI system, which can then echo them in future answers. For example, a story you write in a chatbot may be studied by the company to make the next version of the bot.
- opt out
- Something is on by default and stays on until you find the setting and turn it off. For example, your chats may be used for training unless you go into settings and switch it off.
- personal data
- Any information that is about you or can be linked to you, from your name to your phone's ID. For example, your email, your IP address and the list of apps on your phone are all personal data.
- service providers
- Outside companies that handle your data on the app's instructions, such as a cloud host or an email sender. For example, the company that stores the app's files in the cloud sees your data but only does what the app tells it.
- indemnify
- You promise to pay the company's legal costs and losses if your use of the service gets it sued. For example, if you post a song you do not own and the label sues the app, you owe the app's lawyer bills.
- biometric
- Measurements of your body that identify you, like your face, fingerprint or voice. For example, a selfie used to unlock the app or a voice recording matched to you.
- governing law
- Which country's or state's laws are used to settle any dispute, no matter where you live. For example, a user in Ohio may find the terms are judged under California law.
- sell or share
- Under California law, selling means passing your data to others for money or other value; sharing means passing it on for targeted ads. For example, letting an ad network use your browsing history in return for ad space counts as a sale.
How we got here · grade B · score 50/100 · 15 of 15 policy answers backed by a verified quote · 1 not stated
What does the app collect beyond what it needs to work?
They collect account details, your chats and uploads, payment info, and technical data such as device and advertising IDs, IP-based location, and how you use the site.
Does it record your voice, face or body, and what happens to that?
If you choose age or ID verification, they may collect a photo or video and facial geometry templates linked to that check. Everyday chat voice is not described.
Are your chats and uploads used to train AI models, and is that off by default?
Chats and uploads may train models unless you opt out in account settings. Safety-flagged chats and feedback can still be used after you opt out.
Can employees or contractors read your conversations, and when?
They describe human review for safety, policy, and some third-party content, not a routine sample of all chats with no opt-out.
Are you profiled or tracked for advertising?
They use cookies to market products and let you opt out of sharing for targeted ads of their own services. They also collect advertising identifiers.
Do they sell or share your data, and can you opt out?
They say they do not sell personal data. They share with affiliates, service providers, and connected apps you enable.
What rights do they take over what you type and what the AI makes?
You keep rights in what you submit, and they assign you their rights in outputs, while still using materials to run and improve the service.
How long do they keep your data after you delete it, and can you delete it in the app?
You can delete conversations in the product; they say those are removed from the back-end within 30 days. Broader deletion is by request.
Does it build a lasting memory or profile of you, and can you see, edit or turn it off?
Not stated in the documents.
Do they commit to basic security, and have they leaked data?
They say they use technical and organizational measures to protect personal data, without naming an independent audit in these pages.
Is there a real age gate, and are teens protected?
You must be 18 or older. They say the service is not for children under 18 and they may ask for age checks in some cases.
Will they tell you when the rules change, and is the policy specific?
The privacy policy says they will notify you of material changes. The terms say that keeping using the service after they post updates means you agree. No effective date appears in the text we read.
Can they close your account without warning, and can you get your data out first?
They may suspend or end access without notice if they believe you broke the terms, and they may delete your materials when the account ends.
If something goes wrong, who pays?
The service is as-is, liability is capped at what you paid in six months or $100, and you must cover their legal costs in many disputes.
Do subscriptions renew on their own, and can you get a refund?
Paid plans auto-renew until you cancel at least 24 hours before the next term. They say payments are non-refundable except where the law or a short cooling-off period says otherwise.
If this app can act for you, does it warn you about scams?
May transact on unsafe websites on your behalf. Anthropic asks before some sensitive actions and does not say Claude refuses an unsafe site. This line is not from the privacy policy. We did not test the product. Use Claude in Chrome safely (2026)
The privacy policy is 5,845 words at a professional or legal expert level (Flesch reading ease 0.5); the terms are 4,540 words at a college graduate level.
