Privacy Policy

Your privacy matters to us. This policy explains how Agentic News collects, uses, and protects your information.

Effective: November 6, 2026

Last Updated: November 6, 2026

Quick Summary

Agentic News is a news alerts and interpretation platform that watches the web for topics you care about. To deliver tailored alerts, we ingest licensed and public news sources, de-duplicate and cluster content, and generate role-aware summaries (“Lenses”) with citations. We store article text so our AI can analyze it, and we show you summaries with links back to the source rather than republishing full articles. We do not use your data to train public AI models.

1. Who We Are & Scope

This Privacy Policy explains how Agentic News (“Agentic News,” “we,” “us,” or “our”) collects, uses, shares, and safeguards information about you when you use our websites, applications, and services (collectively, the “Service”). This Policy applies to the Service operated by Agentic News. We act as a data controller for user account data and related usage data.

2. Information We Collect

We collect the following categories of information to operate and improve the Service:

A. Identity & Account Information

  • Email address, and your name when Google or Apple sign-in provides it
  • Authentication identifiers (e.g., Supabase user ID and session tokens)
  • Billing identifiers (Stripe customer ID, subscription status; we do not store full payment card numbers)

B. Service Configuration & Content You Provide

  • Watches (entities/topics you track)
  • Lenses (your role-aware settings)
  • Collections (saved clusters)
  • Explicit feedback (e.g., thumbs up/down, votes on stories in your daily email, survey answers)
  • Chat history with the AI about your news collections
  • Voice recordings, if you dictate an agent description instead of typing it
  • Your LinkedIn profile, if you choose to import it during setup
  • Company websites you ask us to track for changes
  • A Slack webhook address, if you send your daily briefing to Slack, and any webhook addresses you set up to receive story matches
  • Messages you send to support

C. Technical, Usage & Analytics Data

  • Read/open status for items, feature usage timestamps, and API/LLM token consumption
  • Device, browser, and network metadata (“Activity Logs”)
  • IP address (for security, fraud prevention, and coarse geo reporting like country)
  • Product analytics via Amplitude (page views, feature usage, form interactions, your email address and plan once you sign in)
  • Session replays: recordings of how pages were used, made by Sentry (with on-screen text masked and media blocked) on a sample of visits and on visits where an error occurs, and by Amplitude on a sample of visits
  • Whether you open our emails (time and browser) and which links you click (time, IP address, and browser)
  • Your approximate location (city and country) during setup, looked up from your IP address by ipapi.co
  • How you arrived: campaign tags (UTM), the referring site, and an experiment ID that keeps you in the same version of a page

D. Content We Ingest to Power Alerts

  • News articles and public web content used to match your Watches
  • Clustered related items and generated Lenses (summaries with citations)
  • We store the full text of articles we ingest so our AI can analyze them; we show summaries and links back, and do not republish full articles

E. API & Programmatic Access Data

  • API keys (stored as SHA-256 hashes; we cannot retrieve the original key after creation)
  • API request logs (endpoint, timestamp, response status, IP address)
  • MCP (Model Context Protocol) tool call logs (tool name, timestamp, user ID) for usage analytics and abuse prevention
  • OAuth tokens and session data when connecting via third-party AI assistants (e.g., Claude.ai, ChatGPT)

3. How We Use Information

We process your information for:

Providing the Service

Account creation, authentication, matching Watches to clusters, generating Lenses, delivering alerts and digests.Legal bases: contract performance (EU GDPR Art. 6(1)(b)) and/or legitimate interests.

Personalizing & Improving Relevance

Using your feedback and preferences (e.g., thumbs up/down). This tuning is scoped to your experience and internal ranking, not to training public foundation models.Legal bases: contract performance and legitimate interests.

Measuring Usage & Reliability

Enforcing limits (token accounting), preventing abuse, and debugging.Legal bases: legitimate interests.

Billing & Payments

Processing via Stripe; fraud prevention.Legal bases: contract performance and legal obligations.

Communications

Transactional messages about the Service, and where permitted, product updates and marketing (you can opt out).Legal bases: contract performance, legitimate interests, and/or consent where required.

We do not use user data to train public foundation models.

4. Use of AI & Third-Party LLMs

To run the Service we send content to OpenAI: article text, your agents and Lens settings, your chats about your collections, voice recordings you dictate (for transcription), and your LinkedIn profile if you import it. Outputs are returned to the Service for your use, with links to sources.

We also use Anthropic to check the quality of our results internally. Those checks can include your agents' names and descriptions and the stories matched to them.

API & MCP Access by AI Assistants

When you connect an AI assistant (e.g., Claude, ChatGPT) to Agentic News via our API or MCP server, the assistant may send queries and receive your briefing data, story details, and agent configurations on your behalf. We log tool calls for usage analytics and abuse prevention. Your data is processed under the same protections described in this policy — we do not share your personal briefing data with other users or use it to train public AI models.

5. Cookies, Local Storage & Similar Technologies

Cookies

CookiePurposeType
sb-*-auth-tokenKeeps you signed in (Supabase)Essential
an_signin_ticket, an_signin_sentEmailed-code sign-in, 15 minutesEssential
Stripe cookies (e.g. __stripe_mid)Fraud prevention by Stripe, on checkout, billing, and upgrade promptsEssential
an_utm, an_first_touchHow you arrived (campaign tags, referring site), 30 daysAnalytics
Amplitude cookies (e.g. AMP_*)Amplitude device and session IDAnalytics

Local/Session Storage (main categories, with examples)

KeyPurpose
themeLight or dark mode
exp_visitor_id, exp_*A random ID that keeps you in the same version of a page we are testing
an_utm_attribution, an_ai_referrer, an_share_referrerHow you arrived, for the current visit
outreach_refThe referral link you arrived from
Sign-in, setup, and checkout progressCarries you through sign-in, onboarding, and checkout steps
Interface state (e.g. feedSelections, dismissed banners, scroll position)Remembers what you dismissed or selected
Sentry and Amplitude session replayTemporary storage for the session replays described in section 2

You can manage cookies via your browser settings. Blocking essential cookies may impair the Service.

6. Sub-processors & Data Sharing

We use vetted service providers to operate the Service:

ServicePurposeData Shared
Render.comHosting & infrastructureApplication data and server logs
CloudflareNetwork edge for our APIAPI requests
Neon.techManaged PostgreSQLUser data, news metadata, vectors
SupabaseAuthenticationEmail, password if you set one, sessions
OpenAIAI processing and voice transcriptionArticle text, agent and Lens settings, chats, voice recordings, imported LinkedIn profiles
AnthropicInternal quality checksAgent names and descriptions, matched stories
NewsAPI.ai, NewsAPI.org, Hacker News, BlueskyNews sourcesSearch queries built from your agents' topics
FirecrawlWebsite change trackingAddresses of sites you ask us to track
ScrapingdogLinkedIn profile importThe LinkedIn profile identifier from the address you submit, if you choose to import it
ipapi.coApproximate location during setupYour IP address
StripePaymentsBilling info, email
SentryError trackingErrors, stack traces, user ID, session replays with text masked
SendGridEmail deliveryEmail address, digest content, email opens
AmplitudeProduct analyticsUsage events, user ID, email, IP address (for geo), session replays
SlackDaily briefing delivery, only if you connect itYour daily briefing, sent to your own workspace
Your webhooksStory-match delivery, only if you set one upMatched stories, sent to the address you choose

Two public pages show figures added up over many readers' agents. Our transparency page shows counts of recent verdicts on story matches and the headlines and publishers of some rejected stories. For recent days that meet our publishing thresholds, our daily record shows counts of kept and rejected story matches, and the headlines and publishers of stories that at least 3 readers had the same verdict on, with links on the kept headlines. Neither page shows your name, your agents' names, the topics you set up, or the judge's reasons.

We may also disclose information to comply with law, enforce our terms, or protect rights, safety, and the integrity of the Service.

7. International Data Transfers

Where data moves outside your home jurisdiction, we rely on appropriate safeguards (e.g., Standard Contractual Clauses for EEA/UK transfers) and technical/organizational measures consistent with industry practice.

8. Data Retention

Account Data & Preferences

Retained while your account is active. Records of email opens and link clicks are kept until you delete your account.

News Metadata/Cached Content

Retained for archival search. You may email us to request removal of personal data tied to a specific source article.

Application Logs

We generate server logs for security and debugging.

You can delete your account in settings once any paid subscription is canceled. Deletion immediately removes your account, agents, Lenses, collections, feedback, and email history from our database and from our sign-in provider. Copies held by our service providers (for example Stripe billing records, analytics and error events, and email delivery logs) are not deleted automatically; you may email us to request their removal.

9. Your Rights & Choices

Depending on your location, you may have rights to:

  • Delete your account yourself, in settings
  • Request access to, correction of, or a copy of your personal data, by email
  • Object to or restrict certain processing
  • Withdraw consent (where processing relies on consent)
  • Opt out of marketing communications

US State Privacy Rights

California/US state law rights (CA, CO, CT, UT, VA, etc.), including the right to know, delete, correct, and to opt out of “sale” or “sharing” for cross-context behavioral advertising. We do not “sell” personal information as defined by applicable law. Our site loads no ad tags, so we do not “share” personal information for cross-context behavioral advertising.

To exercise rights, contact privacy@agentic-news.ai. We will verify your request and respond within the timelines required by applicable law.

10. Security

We use technical and organizational measures to protect information, including request logging. API keys and AI-assistant access tokens are stored as SHA-256 hashes and cannot be retrieved after creation. AI-assistant connections require PKCE. No method is perfectly secure; please use a strong, unique password and keep API keys and other credentials confidential.

11. Children's Privacy

The Service is not directed to children under 13 (or 16 where applicable law sets a higher age). We do not knowingly collect personal data from children. If you believe a child has provided personal data, contact us to request deletion.

13. Changes to This Policy

We may update this Policy to reflect changes to our practices or legal requirements. We will post the updated Policy with a new “Last updated” date and, where required, provide additional notice.

14. Contact Us

Related Documents