Closeness Lab

Sex Toy App Permissions: A Four-App Snapshot

On 5 August 2026 we opened the US Google Play listings for four companion apps and read the permissions each declares. All four list camera, microphone and precise-location permissions. Two of the four also list calendar read-and-write permissions, and the write permission is described by Google in these words: “add or modify calendar events and send email to guests without owners’ knowledge.”

Research-based, not hands-on (how we review) · This page carries no affiliate links, and nothing below was shown to any manufacturer before publication · Lovense Remote declares the largest permission set of the four and is named throughout; we may earn a commission on Lovense products through our app-controlled vibrator guide (disclosure)

Quick take: this article compares Surfshark’s published 2024 permission dataset with a dated Google Play snapshot for four companion apps. The two headline counts are different measurements, not evidence that an app has become more private. The useful comparison is at the level of individual entries, with the gaps and limits kept visible. Calendar and account access are worth asking about, but a listing does not establish what an app actually does.

The permission descriptions were collected on 5 August 2026. The 18 October revision clarifies the comparison and its limits; it is not a new permission extraction or a test of updated app versions.

What four apps declare, 5 August 2026

Counts are of distinct permission descriptions shown on each app’s US English Play listing. The listing date is the app’s own “Updated on” stamp.

AppCountApp updatedPackage and source
Lovense Remote305 Aug 2026com.lovense.wear (Play listing, retrieved 5 Aug 2026)
FeelConnect 3.0223 Aug 2026com.feeltechnology.feelconnect3 (Play listing, retrieved 5 Aug 2026)
Satisfyer Connect1820 May 2026com.satisfyer.connect (Play listing, retrieved 5 Aug 2026)
We-Vibe, formerly We-Connect1714 Jul 2026com.standardinnovation.weconnect (Play listing, retrieved 5 Aug 2026)

Fourteen descriptions appear on all four listings. Six are the Bluetooth and networking machinery you would expect from a device that pairs with a phone and relays over the internet: pair with Bluetooth devices, access Bluetooth settings, view Wi-Fi connections, view network connections, full network access, receive data from Internet. Five more are housekeeping: control vibration, change your audio settings, prevent device from sleeping, reading the contents of USB storage, and modifying or deleting them. The remaining three are the ones worth naming, because they are on every app in the set: take pictures and videos, record audio, and precise location by GPS and network. Approximate location is on three of the four; We-Vibe’s app declares only the precise kind.

Beyond the shared fourteen, the sets diverge in a way that follows the counts. Satisfyer Connect and We-Vibe declare nothing the others do not. FeelConnect adds one of its own, downloading files without notification. Lovense Remote adds seven: find accounts on the device, create accounts and set passwords, retrieve running apps, modify system settings, control flashlight, allow Wi-Fi multicast reception, and a Google Play license check.

And the calendar pair — reading events including confidential information, and adding or modifying events and emailing guests — is declared by Lovense Remote and FeelConnect, and by neither of the other two. (Two of these apps are listed under names their buyers may not recognise: com.standardinnovation.weconnect is titled We-Vibe on Play and was long known as We-Connect, and com.feeltechnology.feelconnect3 is FeelConnect 3.0. We use the short forms below.)

Comparing the 2024 study with the later Play snapshot

Surfshark’s analysis of ten adult toy apps was published on 13 February 2024. The comparison here uses its linked dataset; the publication date alone does not establish when the page was last revised. Its findings are specific. Across the ten apps it found over sixty permission types, an average of 22 permissions per app and 6 classed as dangerous; Lovense Remote requested the most at 53 permissions and 13 dangerous ones, Satisfyer Connect and Love Spouse 28 each, and NaughtyKit the fewest at 6 and 2. Storage read and write showed up in 8 of the 10, camera in 7, and audio capture in 5 (Surfshark, 13 Feb 2024).

A comparison starts with the methods rather than a subtraction. Surfshark names Exodus Privacy’s public reports as its extraction source and publishes its working dataset as a spreadsheet (Surfshark research dataset, retrieved 5 Aug 2026). We downloaded it on 5 August 2026 and reopened the currently linked sheet on 17 October for this clarification. We do not have its version history, so the study’s publication date does not prove that the file is unchanged from its first release. References to the study’s sheet below identify that linked source, not an independently archived original. It lists all ten apps by Play URL, and three of those package identifiers are three of our four: com.lovense.wear, com.satisfyer.connect and com.standardinnovation.weconnect.

Two things do not transfer, and both have to be said before any comparison means anything.

The first is the headline number. Surfshark counted permission names declared inside the app; the public Play listing shows Google’s human-readable descriptions, where several underlying permissions collapse into a single line or into none. Their 53 for Lovense Remote and our 30 are not the same measurement, and subtracting one from the other produces a number that means nothing.

The second is the wording. Surfshark’s sheet pairs manifest names with descriptions in cells such as Dangerous - CAMERA take pictures and videos. The descriptions differ between the sources: the sheet has “read the contents of your shared storage”, while our Play snapshot has “read the contents of your USB storage”; the other pairs are “have full network access” and “full network access”, “prevent phone from sleeping” and “prevent device from sleeping”, and “access approximate location only in the foreground” and “approximate location (network-based)”. This establishes a wording difference, not who changed a string or when. Exact text matching can miss correspondences; any mapping from a description to a manifest name needs checking rather than treating different wording as a permission change.

There is a third problem, and it decides what can honestly be claimed in either direction. An item is comparable only if its manifest name carries a human-readable description in the source sheet — and Surfshark’s sheet lists plenty that did not. Twenty-two of Lovense Remote’s fifty-three entries are a bare name with no description: FLASHLIGHT, CHECK_LICENSE, RECEIVE, AD_ID, POST_NOTIFICATIONS and so on. Seven of Satisfyer Connect’s twenty-eight and seven of We-Vibe’s twenty-six are the same.

That trap is live in the table above. Lovense Remote’s unique descriptions in our 5 August snapshot include “control flashlight” and a Google Play license check. Both look like recent additions and neither is: FLASHLIGHT and CHECK_LICENSE are both in its row in the linked sheet, just without a description attached. Our own data caught us on that one.

These checks identify differences worth following up, not proved changes to app code. The sources use different methods, and app builds, Android versions and store presentation can affect what appears. BLUETOOTH_CONNECT and BLUETOOTH_SCAN, for example, are in the 2024 data for all three matched apps but have no corresponding description in any of the four later listings we read. That absence does not prove the apps stopped using Bluetooth. A description still shown for another app is a useful cross-check, but it does not turn a missing store entry into proof that an app removed a permission.

With those limits, these are the entry-level differences across the three package IDs present in both sources. All later-listing observations refer to 5 August 2026.

Calendar entries appear in the later Lovense Remote listing. No calendar permission appears anywhere in the sheet linked from the 2024 study, for any of the ten apps, described or not. com.lovense.wear in our later snapshot declares both reading calendar events including confidential information and adding or modifying them.

Account entries appear there too. On the same comparison: the study’s linked sheet contains no account permission of any kind for any of the ten apps. Our Lovense Remote snapshot declares both finding accounts on the device and creating accounts and setting passwords, and it is the only one of our four that declares either. The next section considers possible feature context, without claiming to establish which feature uses each permission.

Satisfyer Connect has location entries in the later listing. In the linked sheet it is the only one of the ten apps declaring no location permission of any kind. Its later listing includes both approximate and precise location descriptions.

We-Vibe’s later listing includes “read phone status and identity”. The corresponding permission is marked dangerous in the study’s linked sheet, where it is present for Lovense Remote but not for the other two matched apps.

Two smaller listing differences. “reorder running apps” appears in the later Lovense Remote and Satisfyer Connect snapshots; its entry in the linked sheet was absent for Satisfyer and present for We-Vibe. “approximate location (network-based)” appears in three of our four later listings, but not We-Vibe’s, whose snapshot lists precise location. These are differences between declarations, not verified removals from the installed apps.

And one finding in the linked sheet this method cannot re-check at all: Lovense Remote was the only app of the ten declaring background location, and the descriptions in our Play snapshots do not distinguish background from foreground, so they do not settle whether that access remains.

Counted under that matching rule, eight described entries appear in the later snapshots without a corresponding declared entry in the study’s linked sheet, and two older entries have no description in the later We-Vibe snapshot. That is a comparison of the two sources, not a verified eight additions and two removals in app code. One further description, “read Google service configuration”, is absent from the study’s linked sheet and appears on two later listings; it is recorded here separately rather than silently omitted. Establishing actual permission changes would require comparable app builds and a separate technical audit.

Feature descriptions give context, not a behaviour test

It would be easy to write the camera and the microphone up as sinister, and it would be wrong. Lovense’s Play description gives possible context. Media Sync “let[s] music, sounds, or videos drive your toy’s vibrations”, a feature relevant to audio access. Burn After Reading lets photos and videos sent in the app expire, which gives context for camera access without establishing that every use requires it. The listing also advertises Control Link and Control Roulette, features for handing control to another person or to a stranger for a limited time, which may help explain account and networking requests but does not establish which permission each feature uses (Lovense Remote Play listing, retrieved 5 Aug 2026).

A feature list gives you questions to ask before granting access, not a guarantee about the result of denying it. Review permissions against the functions you intend to use. We have not tested each app with permissions denied, so we cannot promise that refusing camera, microphone or location access leaves pairing and every other function unaffected.

The calendar entries on Lovense Remote and FeelConnect are the ones we could not connect to an explanation in the listings we read. An embedded library is one possible explanation, not an established cause; we have not inspected the apps to identify it. The declared access is a question to put to the developer rather than evidence that the app reads a calendar in practice.

What permissions do not tell you

Denying the camera stops the app reaching the camera. It does not stop the app recording that you opened it, which toy you paired, which pattern you chose or how long the session ran, because none of that comes from your phone’s hardware — it is the app’s own record of its own use. That is governed by a different declaration, the Play data safety section, and we went through what four brands say there, including three places a company disagrees with its own filings, in what sex toy apps say they collect.

The written policies do not close the gap either. A content analysis published in April 2026 examined the privacy policy documentation and product descriptions of 146 smart sex toys across fourteen brands, and reported that among the policies available, critical data storage and transmission details were “frequently omitted or ambiguous”, leaving them misaligned with the privacy claims made in the same companies’ marketing (Sun, Gómez Ortega, Rostami and Lampinen, CHI 2026). We are quoting the authors’ own summary of their findings rather than a reading of the full paper, and saying so.

So the picture is three declarations that do not reconcile: what the app can reach, what the company says it collects, and what the policy says it does with it. The permission list is the only one of the three you can verify in a minute, which is a reason to look at it and not a reason to think it settles anything.

How to check any app yourself, in about a minute

If you have not bought yet, the prior question is whether you want an app-connected device at all, which we put first in our guide to choosing a couples toy. If the security record is what you are weighing, the dated list of what has actually happened is in can sex toys be hacked.

Limits of this snapshot

Four apps is four apps, and the four were chosen because this site already covers those brands, not by any sampling method. Surfshark looked at ten and found a wider spread than we did.

These are Android listings for a US English store. Play listings vary by region, iOS reports permissions differently, and a company can change what it declares with any update; one of these four was updated the same day we read it. One counting trap, since the numbers are the point: most of these descriptions begin with a lower-case letter and one does not (“Google Play license check”), so a filter that assumes lower case silently loses it. Our first pass did exactly that and reported 29 for Lovense Remote instead of 30. Everything above is a snapshot with a date on it, which is why the date is in the table and the method is written out. If you repeat it and get different numbers, check the date, region, app version and counting rule before inferring an app change. That is the point of publishing the method rather than only the result.

We have not decompiled any app, inspected any traffic, or verified that any app’s behaviour matches what it declares, and nothing above rests on our having done so.

FAQ

What permissions do sex toy apps ask for? We read the US Google Play listings for four companion apps on 5 August 2026 and counted the distinct permission descriptions each declares: Lovense Remote 30, FeelConnect 3.0 22, Satisfyer Connect 18, and the app listed as We-Vibe (formerly We-Connect) 17. Fourteen descriptions are common to all four, and among them are taking pictures and videos, recording audio, and precise location by GPS and network. Two of the four, Lovense Remote and FeelConnect, also declare reading calendar events including confidential information, and adding or modifying calendar events and sending email to guests without owners’ knowledge.

Why does a vibrator app need my camera and microphone? Media and social features give possible context: Lovense advertises Media Sync and Burn After Reading. But a listing is not a permission-by-feature test, and we have not tested which functions keep working when access is refused. Review each request against the features you intend to use; do not assume every request is necessary or that denying it has no functional cost.

Are the numbers in the Surfshark study still accurate? Surfshark published its study on 13 February 2024 and linked the dataset used here. We compared it with a 5 August 2026 Google Play snapshot, not a new extraction using the same method. Exodus permission names and Play descriptions are different units, so their 53 for Lovense Remote and our 30 cannot be subtracted. Three of our four package IDs match its dataset. Calendar, account and other entry-level differences are worth checking, but do not by themselves prove changes in app code or behaviour.

Does denying permissions make the app private? It closes one door of two. Denying the camera stops the app reaching the camera; it does not stop the app recording that you opened it, which toy you paired, or how long the session ran, because that is the app’s own data rather than your phone’s. Permissions govern what an app can reach on the device. What a company says it collects is a separate declaration, in the Play data safety section, and we went through what four brands declare there in a companion piece.

Related reading: What sex toy apps say they collect → · Are sex toy apps safe? → · Can sex toys be hacked? → · How to choose a couples sex toy → · How we review →

Sourcing: the permission counts were extracted by us from the four Play listings linked in the table on 5 August 2026, and the method is written out above so the count can be repeated or corrected. Store listings can be changed by the developer at any time and vary by region. For adults 18+. General information, not security or legal advice.