What information can a website collect about me without permission?

Without a single prompt, a website can read your IP address, approximate city, device type, operating system, browser and version, screen resolution, battery level, installed fonts, GPU, connected cameras and microphones, and a canvas/audio fingerprint. This page demonstrates dozens of these signals live in your own browser.

What is a browser fingerprint?

A browser fingerprint is a near-unique identifier assembled from your device's characteristics — screen size, fonts, GPU, timezone, and more. The Electronic Frontier Foundation found that only about 1 in 286,000 browsers share the same fingerprint, so most people can be recognised across sites without any cookies at all.

Can a website see my exact location?

By default a site only sees your approximate city, derived from your IP address. Pinpoint GPS location — accurate to a few metres — requires your explicit permission, but many apps and sites request it as a matter of routine.

Can websites see my other tabs or which sites I'm logged into?

No. The browser sandbox and same-origin policy stop any page from reading your other tabs, your history, or another site's data. However, a page can still detect the moment you switch away from its tab, the site you arrived from, and — sometimes — which major services still recognise your browser session.

How do I stop websites from tracking me?

Install a content blocker such as uBlock Origin, use a privacy-focused browser (Brave, Firefox, or Tor), deny camera, microphone, location, and clipboard access by default, block third-party cookies, and turn on Global Privacy Control. Together these steps cut off the large majority of trackers.

Does this website store any of my data?

No. Every reading is computed inside your browser and never leaves it, apart from one optional lookup that turns your IP address into an approximate city. Visits are measured only with anonymous, cookieless counts that cannot identify you.

Is my phone really listening to my conversations for ads?

Almost certainly not in the way people fear. The largest independent studies found no evidence that mainstream apps secretly record conversations and use them for ad targeting. The 'creepy' ads are far better explained by profiling, shared networks and accounts, location data, retargeting, and the fact that you only notice the ads that hit.

So no device has ever used the microphone to track people?

Some have — but not to record your conversations for ads. Software like Alphonso used the mic to detect what you watched on TV, some smart TVs captured viewing and voice data, and voice assistants send audio to servers (including accidental wake-word triggers, which humans have reviewed). These are real, documented cases, and most were disclosed in a privacy policy.

How can I tell if an app is actually using my microphone?

Check the microphone indicator dot (orange on iPhone, green on Android) that appears whenever the mic is active, and review which apps hold microphone permission in your phone's privacy settings. Revoke access for anything that doesn't need it, and the mic simply can't be used by that app.

Is My Phone Listening to Me for Ads? What the Evidence Says

Almost everyone has a story: you talk about a product you've never searched for, and an ad appears hours later. It feels like proof your phone is secretly listening. The best available evidence says it usually isn't — the truth is more mundane and, in a way, more unsettling: advertisers rarely need your microphone, because everything else about you already gives them the answer.

Updated July 2, 2026 · 8 min read

The short answer

There is no credible technical proof that mainstream apps secretly turn on your microphone, transcribe your conversations, and use them to target ads. Multiple independent teams have gone looking for exactly that behaviour and failed to find it.

That does not mean nothing is listening. A handful of real, documented products have used device microphones — but for narrower purposes than 'record everything you say', and usually buried in a privacy policy you agreed to. And voice assistants genuinely do send audio to servers, especially when they mishear their wake word.

What the research actually found

In 2018, a team at Northeastern University ran the largest test of this claim to date — a project nicknamed 'Panoptispy'. They automatically analysed more than 17,000 popular Android apps, watching for any that activated the microphone and sent audio off the device. They found no evidence of apps secretly recording conversations for ads.

They did, however, find something else: some apps were quietly recording the screen — screenshots and video of what users did — and sending it to third-party analytics companies. In one case a food-delivery app leaked recordings that could include personal information. So the popular fear was unfounded, but the instinct that 'apps are watching' wasn't entirely wrong; they were just watching a different channel.

Separate tests by security firms reached the same conclusion about audio. Leaving phones in a room with constant talking versus silence produced no meaningful difference in data uploaded or battery drained — the tell-tale signs continuous audio surveillance would leave behind.

Real cases where devices genuinely did listen

Microphone-based tracking is not science fiction. It has happened — just not in the 'listening to your conversations for ads' form people imagine:

Alphonso (2017): The New York Times reported that software from a company called Alphonso was embedded in hundreds of mobile games — some aimed at children — and used the phone's microphone to listen for audio signatures from TV shows and ads, so advertisers could tell what you watched. It was disclosed in the apps' descriptions, but easy to miss.
Smart TVs — Vizio (2017): The FTC and New Jersey fined Vizio $2.2 million for capturing second-by-second viewing data from 11 million televisions using automatic content recognition, without clearly telling owners, then selling it to advertisers.
Samsung smart TVs (2015): Samsung's own privacy policy warned that its voice-recognition feature could capture spoken words — including personal or sensitive information — and transmit them to a third-party service. The line 'don't discuss personal information in front of your TV' became infamous.
Voice assistants (2019): Reporting revealed that human contractors were reviewing snippets of Alexa, Siri, and Google Assistant recordings to improve accuracy — including clips captured by accidental 'wake word' triggers. Apple apologised and made review opt-in; it later agreed to a $95 million settlement over Siri recordings. Amazon paid $25 million to the FTC in 2023 over how long it retained children's Alexa recordings.

The 'Active Listening' pitch that fuelled the panic

In late 2024, journalists at 404 Media obtained a marketing deck from Cox Media Group (CMG) promoting a product it called 'Active Listening'. The pitch claimed CMG could use AI to capture voice data from smart-device microphones and use it for ad targeting — and it named Google, Amazon, and Facebook among its partners.

The reaction was swift. Google removed CMG from its Partner Program, Meta said it would review whether CMG had breached its terms, and Amazon stated it had never worked with CMG on such a product. The consensus among researchers was that this was an advertising sales claim — overselling what is technically and legally feasible at scale — rather than proof that the big platforms secretly listen. But it captures exactly why the myth persists: some companies actively market the fantasy, even if they can't deliver it.

Why the ads feel so accurate anyway

If nobody's listening, why does it happen so often? Because advertisers already know enough to predict what you'll talk about, and because your brain notices the hits and forgets the misses.

Shared context: the person you spoke with searched for it, and you share a Wi-Fi network, IP address, location, or are linked as 'people who are often near each other' in ad-platform data.
Rich profiling: your past searches, purchases, app usage, and browsing already model your interests well enough to surface something you were about to want anyway.
Location and timing: visiting a shop or a friend's house is logged, and ads for related products follow that movement.
Retargeting: you (or someone on your account) glanced at the product days ago, and the ad simply resurfaced when you'd half-forgotten.
The frequency illusion: once a topic is on your mind you suddenly notice ads for it that were always there — and you never count the thousands of irrelevant ads you scrolled past.

How to check and shut it down

Review microphone permissions: on iOS, Settings → Privacy & Security → Microphone; on Android, Settings → Privacy → Permission manager → Microphone. Revoke access for any app that has no reason to need it.
Watch the indicator dot: modern iPhones and Android phones show an orange/green dot when the mic is active. If it lights up when no app should be listening, investigate.
Rein in your voice assistant: turn off human review of recordings, delete stored voice history, and consider disabling 'always listening' wake words.
Attack the real pipeline: the tracking that actually drives creepy ads is the profiling one. Turn off ad personalisation in your Google, Meta, and TikTok settings, reset your mobile advertising ID, and use a content blocker like uBlock Origin on the web.

Frequently asked questions

Is my phone really listening to my conversations for ads?: Almost certainly not in the way people fear. The largest independent studies found no evidence that mainstream apps secretly record conversations and use them for ad targeting. The 'creepy' ads are far better explained by profiling, shared networks and accounts, location data, retargeting, and the fact that you only notice the ads that hit.
So no device has ever used the microphone to track people?: Some have — but not to record your conversations for ads. Software like Alphonso used the mic to detect what you watched on TV, some smart TVs captured viewing and voice data, and voice assistants send audio to servers (including accidental wake-word triggers, which humans have reviewed). These are real, documented cases, and most were disclosed in a privacy policy.
How can I tell if an app is actually using my microphone?: Check the microphone indicator dot (orange on iPhone, green on Android) that appears whenever the mic is active, and review which apps hold microphone permission in your phone's privacy settings. Revoke access for anything that doesn't need it, and the mic simply can't be used by that app.