Privacy policy

This page covers Folio by Zip1 at www.zip1.ai and the Folio app for iPhone, which shows this same site. Folio is run by [LEGAL ENTITY NAME]. Questions go to [CONTACT EMAIL].

In short

• There are no accounts. Folio never asks for your name or email address.
• Folio’s code sets no cookies and includes no analytics or tracking scripts.
• When you search, your search text goes to Folio’s server. The server passes it to the sources you have on to get results. Folio does not save your searches in a database.
• AI answers are off unless you turn them on. Only then does your search, with the top results, go to the AI provider you picked (xAI, OpenAI, or Anthropic).
• Your source settings and recent searches are kept in your browser, not on our server.
• Ads are off unless you turn them on. Even then, your search is never sent to the ad network.

What you type and where it goes

A search sends what you typed (up to 180 characters) to Folio’s server, along with which sources are on and which page of results you want. If a search address includes a place (the “near” part of the address), that place is sent too. The server then asks these services for results:

• Microsoft Bing: web results, related “Deep dive” searches, and image results when Images is on.
• Wikipedia (Wikimedia Foundation): articles, the summary card, and coordinates for place names, when Wikipedia is on.
• Grokipedia (xAI): pages and summaries, when Grokipedia is on.
• One AI provider, only when AI is on: xAI for Grok, OpenAI for ChatGPT, or Anthropic for Claude, whichever you picked. It receives your search plus the titles, snippets, and addresses of the top results from the sources that are on (for Images, each picture’s title and the page it was found on), to write the AI answer. The other two providers receive nothing. If the provider you picked does not answer, Folio sends the same request to the next provider that is set up, and the answer says which one wrote it. Each provider handles the request under its own API terms and privacy policy.
• Datamuse: short definitions for words in your search.
• Open-Meteo geocoding: place lookups for the location references.

As you type, once there are at least two characters, Folio’s server sends the text so far to Wikipedia and Grokipedia for suggestions. When you open a result preview, the server sends that result’s title and address to Wikipedia or Grokipedia to get a longer extract. Folio’s server makes all of these requests itself. They carry the server’s own identification, not your IP address, cookies, or device details. Each service handles the requests it receives under its own policy. Trending topics come from Google Trends and Grokipedia, and Folio’s server fetches them without sending anything about you.

Your search is also part of the page address (for example /?q=…), so it appears in your browser history and in the address you share if you copy the link. The public search endpoint at /api/search works the same way. The query is in the address, and results come from the same services. It sends the query to an AI provider only when the address asks for an AI answer (ai=1).

What stays in your browser

Folio keeps four items in your browser’s local storage for this site:

• which sources you turned on or off;
• which AI provider you picked, if you picked one;
• your last six searches, shown under Recent;
• whether you turned sponsored listings on.

These stay on your device. Folio’s server does not receive them as stored data. It only receives the search you are running. To remove them, use Clear next to Recent, or clear this site’s data in your browser settings.

On Folio’s server

Folio has no user database and does not write your searches to its own logs. To avoid repeating the same requests, the server briefly keeps some source responses in memory (for up to ten minutes, and only the most recent few dozen). That memory is not linked to you and is cleared when the server restarts.

Folio is hosted on Vercel. Like any web host, Vercel receives each request’s IP address, browser details, time, and the address requested, which can include your search text. How long Vercel keeps these request logs is [VERCEL LOG RETENTION PERIOD]. See the Vercel privacy notice.

What your browser loads from other services

Some parts of a Folio page load straight from other companies’ servers. Those companies see your IP address and browser details when they do:

• Google Fonts serves Folio’s typefaces on every page.
• Google’s favicon service serves the small site icons next to results. That request includes the result’s domain name.
• Images in results come from Bing’s image servers, Wikimedia, Google (pictures on trending topics), and, when you open an image preview, the site that hosts that image.
• YouTube (youtube-nocookie.com) loads only when you open the preview of a YouTube result.
• A “Created with Grok” script from grok.com (xAI) loads on each page. It checks with xAI’s app-builder service whether to show a small “Created with Grok · Remix” banner that links to Grok.

Related policies: Google, Microsoft, Wikimedia, xAI, OpenAI, Anthropic.

Ads

Sponsored listings are off by default. If you turn on “Call the ad network” under Ad preferences, Folio’s server asks the Kevel ad network for one ad and reports that it was shown. That request has no search text and no information about you. If an ad appears, its image loads from Kevel’s servers in your browser, and clicking it goes through Kevel’s click link, so Kevel sees your IP address and browser details at that point. How Kevel uses that information: [AD PARTNER DATA USE]. Turn the setting off to stop ads.

Links you open

Results link to other sites, which have their own privacy practices. The arrow on a result opens the page without passing Folio’s address along. In the iPhone app, outside pages open in Apple’s in-app Safari view.

The iPhone app

The Folio app shows this website, so everything above applies. The app itself does not add accounts, analytics, advertising identifiers, or tracking, and it does not ask for your location, photos, or contacts. The Share button hands a link to the iOS share sheet only when you tap it, and you choose where it goes.

What Folio does not do

• Folio does not sell or rent information about you.
• Folio does not build profiles or track you across other sites and apps.
• Folio does not ask for or knowingly collect personal information from anyone, including children.

Results come from the open web. Folio does not add its own content filter, so be aware of this if a child is using it.

Your choices

• Turn Wikipedia, Grokipedia, or Images off so a search skips them.
• Leave AI off so your search is never sent to an AI provider.
• Clear Recent, or clear this site’s data in your browser.
• Leave sponsored listings off.
• Avoid typing personal details into the search box, since searches are sent to the services above.

Changes

If Folio’s handling of data changes, this page and the effective date above will change with it. Questions or requests: [CONTACT EMAIL].