Every day, millions of people type their deepest work problems, their personal anxieties, their business strategies, their medical symptoms, their relationship difficulties, and their financial decisions into AI chatbots. They type them into a box on a screen, press Enter, and receive a response.
Most of them have never read the privacy policy of the tool they are using. Most have never asked what happens to that information after they send it. Most assume, without much thought, that it disappears into the digital ether.
It does not. And the details of what actually happens to your data when you use AI tools are worth understanding clearly, whether you are a student in Edinburgh, a business owner in Vancouver, a professional in Chicago, or a parent in Sydney.
This post gives you the honest, specific answers. What the major AI companies collect, what they do with it, what your rights are, and most importantly, what you should and should not be typing into these tools.
Important Note: This post covers the privacy practices of major AI companies as understood
from their publicly available terms of service and privacy policies as of 2026.
These policies change. Always check the current privacy policy of any tool you
use, particularly before sharing sensitive information.
What These Companies Are Actually Collecting
Your
Conversation Content
The prompts you type and the responses you receive are stored on
the AI company's servers. This is not unique to AI tools. Email providers,
messaging apps, and search engines have always stored communication content.
What is different with AI tools is the nature of what people share.
People routinely share things with AI tools they would share
with very few humans: detailed descriptions of business strategies, personal
health situations, relationship problems, financial circumstances, and
confidential professional information. The AI feels like a private
conversation. It is not.
Usage
Metadata
Beyond conversation content, AI companies collect standard usage
data including your IP address, device information, browser type, geographic
location (approximate), feature usage patterns, timestamps, and session
duration. This data is used for service improvement, security monitoring, and
often for advertising targeting in platforms that monetise through ads.
What Each Major Tool Does with Your Data
ChatGPT
(OpenAI)
By default, OpenAI uses conversations from ChatGPT to train and
improve its models. This means the content you type can, in principle,
influence future versions of the model. OpenAI provides an opt-out mechanism:
users can disable chat history in Settings, which prevents conversations from
being used for training.
When chat history is disabled, conversations are still processed to generate responses but are not retained after 30 days and are not used for training. ChatGPT Plus subscribers have access to the same controls.
OpenAI is based in San Francisco, United States, and processes data primarily in the United States. For users in the United Kingdom and European Union, this raises GDPR cross-border data transfer considerations. OpenAI has implemented Standard Contractual Clauses to address this.
What to do: Go to ChatGPT Settings, then Data Controls, and disable 'Improve
the model for everyone' if you do not want your conversations used for
training. This is the single most important privacy setting in ChatGPT.
Claude
(Anthropic)
Anthropic's privacy approach is generally regarded as more
conservative than OpenAI's. By default, Anthropic states that it does not use
conversations from Claude.ai to train its models without explicit consent,
though conversations may be reviewed by staff for safety and quality purposes.
Anthropic emphasises its safety-focused mission and has designed its data handling with a degree of privacy consideration that many security-conscious professionals find more reassuring than alternatives. This is one reason Claude has gained significant traction among legal and medical professionals in the United States, United Kingdom, and Canada who work with sensitive information.
That said, conversations are still processed and stored, and Anthropic's privacy policy permits data use for a range of service-related purposes. Claude is not a private communications channel.
What to do: Review Anthropic's current privacy policy at anthropic.com
before sharing sensitive professional information. For enterprise use cases
requiring stricter data handling, Anthropic's API and enterprise plans offer
different data retention terms.
Google
Gemini
Google's data practices reflect its status as one of the world's
largest advertising companies. By default, conversations with Gemini are stored
and can be reviewed by human reviewers to improve Google's products and
services. Google states this clearly in its terms.
Google Gemini activity is linked to your Google account, which means it exists alongside your search history, Gmail, YouTube viewing history, and other Google product data within Google's broader data ecosystem. For users in the United Kingdom, Australia, Canada, and New Zealand who use Google Workspace, the enterprise data handling terms differ from consumer terms.
Users can turn off Gemini Apps Activity in their Google Account settings to prevent conversations from being saved to their account and used for model improvement.
What to do: Go to myaccount.google.com, then Data and Privacy, then Gemini
Apps Activity, and turn it off if you do not want conversations stored and
linked to your Google account.
The Five Things You Should Never Type into an AI Tool
1. Passwords and login credentials: This should be self-evident but bears stating: never type a
password, PIN, security question answer, or any authentication credential into
an AI tool under any circumstances.
2. Confidential client or patient
information: If you are a lawyer, doctor, accountant,
or any other professional with a duty of confidentiality, typing client or
patient details into a consumer AI tool almost certainly violates your
professional obligations and potentially applicable law. In the United Kingdom,
this includes obligations under the Solicitors Regulation Authority and the
General Medical Council. In Canada, provincial privacy legislation applies. In
Australia, the Privacy Act 1988 and professional regulatory requirements are
relevant. Enterprise AI plans with specific data processing agreements are
available for professional use.
3. Financial account details: Bank account numbers, credit card numbers, tax file numbers,
Social Security numbers in the United States, National Insurance numbers in the
United Kingdom, and equivalent identifiers in Canada, Australia, and New
Zealand should never be shared with any AI tool.
4. Unreleased business strategies and
competitive information: Information about unannounced products,
acquisition targets, pricing strategies, and other confidential business
intelligence can be processed and stored by AI companies. This is a real
corporate risk that legal and compliance teams at companies in every country
are actively managing.
5. Information about other people
without their knowledge: Typing detailed personal information
about a colleague, family member, or third party into an AI tool without their
knowledge raises both ethical and, in some jurisdictions, legal concerns under
privacy legislation including GDPR in the UK and Europe, PIPEDA in Canada, and
the Privacy Act in Australia.
What Your Legal Rights Actually Are
Your privacy rights when using AI tools depend significantly on where you live. Here is a country-by-country summary of the most relevant protections.
United
States
There is no single comprehensive federal privacy law governing
AI in the United States as of 2026. Privacy protections come from a patchwork
of sector-specific laws and state legislation. California's Consumer Privacy
Act (CCPA) and its successor, the California Privacy Rights Act (CPRA), give
California residents the strongest protections, including the right to know
what data is collected, to request deletion, and to opt out of data sale.
Residents of other states have varying levels of protection under their state
laws.
United
Kingdom
The United Kingdom retained a version of the GDPR as UK GDPR
following Brexit, which gives UK residents robust privacy rights including the
right to access their data, the right to erasure, the right to object to
processing, and the right to data portability. The Information Commissioner's
Office (ICO) is the supervisory authority. UK users can submit data subject
access requests to AI companies operating in the UK.
Canada
Canada's Personal Information Protection and Electronic
Documents Act (PIPEDA) governs how private sector organisations collect, use,
and disclose personal information. Canada is also in the process of updating
its federal privacy framework. Canadian users have the right to know what
personal information an organisation holds about them and to challenge its
accuracy.
Australia
The Privacy Act 1988 and the Australian Privacy Principles
(APPs) govern how organisations handle personal information. The Office of the
Australian Information Commissioner (OAIC) oversees compliance. Australian
users have the right to access personal information held about them and to
request corrections. Australia is currently reviewing and updating its privacy
framework to better address emerging technology including AI.
New
Zealand
The Privacy Act 2020 is New Zealand's primary privacy
legislation, overseen by the Office of the Privacy Commissioner. It includes
information privacy principles governing collection, use, disclosure, and
security of personal information. New Zealanders have the right to access and
correct personal information held about them.
Practical Steps to Protect Your Privacy with AI Tools
1. Disable training data opt-ins. In
ChatGPT, turn off 'Improve the model for everyone' in Settings. In Gemini,
disable Gemini Apps Activity. Check and adjust privacy settings in every AI
tool you use regularly.
2. Read the privacy policy of new tools
before you share sensitive information. This does not mean reading every word.
It means checking what data is collected, how long it is retained, and whether
it is used for training.
3. Use enterprise or business plans for
professional use. Consumer free tiers typically have the broadest data
collection terms. Enterprise plans from OpenAI, Anthropic, and Google offer
more restrictive data handling with contractual commitments.
4. Anonymise where possible. If you need AI
help with something involving real people or real businesses, replace specific
names and identifying details with generic placeholders. The AI does not need
to know the real names to help you effectively.
5. Submit data subject access requests if
you want to know what is held. Under UK GDPR, CCPA, PIPEDA, and other
applicable laws, you have the right to request a copy of personal data held
about you by AI companies operating in your jurisdiction.
6. Stay informed. Privacy policies change.
The AI Vanguard covers significant AI privacy developments in its AI Safety and
Privacy category. Subscribe to stay updated.
The Honest Assessment
The major AI companies are not rogue actors deliberately misusing your data. They are technology businesses operating within their stated terms of service, which most users never read.
The privacy risks of AI tools are real but manageable. The most significant risks come not from AI companies themselves but from users sharing information that should never be shared with any third-party service, confidential client data, personal identifiers, credentials, and commercially sensitive strategy.
The bottom line: use AI tools freely for general tasks. Apply the same judgment you would apply to any cloud service when deciding what to share. And take five minutes to adjust the privacy settings on the tools you use most. That five minutes is the most practical privacy investment most people can make.
Key Takeaways
•
AI
tools collect and store your conversation content and usage metadata. This data
is used in ways that vary by tool and are governed by each company's privacy
policy
•
ChatGPT
uses conversations for model training by default. This can be disabled in
Settings. Claude has more conservative default training practices. Gemini links
activity to your Google account and uses it for product improvement
•
Never
share passwords, confidential client information, financial account details,
unreleased business strategies, or personal details about others without their
knowledge
•
Your
privacy rights depend on your location. UK GDPR, CCPA in California, PIPEDA in
Canada, the Australian Privacy Act, and New Zealand's Privacy Act 2020 all
provide relevant protections
• Practical steps: disable training opt-ins, read privacy policies for new tools, use enterprise plans for professional sensitive work, and anonymise specific details when possible
Frequently Asked Questions
Can AI companies read my conversations?
Yes, in principle. All major AI companies state in their privacy
policies that conversations may be reviewed by employees for safety, quality
assurance, and service improvement purposes. In practice, the vast majority of
conversations are processed by automated systems only. But the technical and
legal capacity for human review exists. This is another reason not to share
genuinely confidential information.
Is ChatGPT HIPAA compliant for medical use?
The standard consumer version of ChatGPT is not HIPAA compliant and should not be used with protected health information. OpenAI's enterprise and API offerings can be configured for HIPAA compliance under a Business Associate Agreement. Healthcare professionals in the United States, Canada, Australia, and the United Kingdom should use only tools and configurations specifically approved for use with patient data by their organisation and relevant regulatory bodies.
Can I request deletion of my data from AI companies?
Yes, in many jurisdictions. Under UK GDPR, the right to erasure
allows UK residents to request deletion of personal data. CCPA gives California
residents similar rights. OpenAI, Anthropic, and Google all have data deletion
processes available through their privacy portals or by contacting their privacy
teams. The effectiveness and completeness of deletion from training data is an
area of ongoing legal and technical debate.
Is using AI at work a privacy risk for my employer?
Potentially, yes. Employees who use consumer AI tools with confidential company information may be violating their employment agreements and their employer's data security policies, as well as potentially applicable data protection law. Many organisations in the United States, United Kingdom, Canada, Australia, and New Zealand have issued AI usage policies. If your organisation has not, it is worth raising the question with your manager or IT security team.
AI Safety and Privacy: The AI Vanguard covers AI privacy, regulation, deepfakes,
responsible use, and the broader safety questions around artificial
intelligence every week. Subscribe below to stay informed.
