Internet Privacy Statistics 2026: Data, Breaches and Consumer Concerns

Written by Nick Galov • Edited by Paul Tsurkan
• 10 min read

Do you realize how much of your personal information is on the internet?

Everyone who uses the internet leaves a trail, and the statistics describing what happens to that trail are unusually easy to misread. The single most common error is treating how people feel about privacy as evidence of what has happened to them.

Being worried about data collection is not the same as having your data stolen. A breach notice is not a victim. An account takeover is not every kind of hacking. So before the numbers, the definitions they rely on:

  • Privacy attitudes are survey measures of concern, trust, understanding or sense of control. They describe opinion, not events, and concern can rise while incidents fall.
  • Account takeover means someone gained control of an account you hold. It is one specific incident type, not a synonym for being hacked.
  • A data compromise (or breach) is one incident at one organisation, however many people it touched.
  • A victim notice is one notification sent to one person about one compromise. A single person breached three times receives three notices, so notices always exceed unique people.
  • Identity theft is the misuse of personal information, which may or may not follow a breach. It is covered separately in our identity theft statistics.
  • A government data request is a demand to a platform for user information. It records that a request was made, not that it was granted or that anyone's privacy was violated.

Every figure below names its source, population and period. Nearly all the attitude data is from Pew Research Center surveys of US adults, which means it describes Americans rather than internet users worldwide.

Key Internet Privacy Statistics for 2026

  • 81% of US adults are concerned about how companies use the data they collect; 71% say the same about government.
  • 73% feel they have little or no control over data companies collect, and 79% feel that way about government data.
  • 72% say they have little or no understanding of the laws protecting their data privacy.
  • 56% frequently click “agree” on privacy policies without reading them.
  • 11% had an email or social media account taken over in the previous 12 months.
  • The ITRC recorded 3,322 US data compromises in 2025, an all-time record.
  • Those compromises produced 278.8 million victim notices, down 79% from 2024 despite the record incident count.
  • 70% of 2025 breach notices gave no information about how the attack happened.

How Americans Feel About Data Privacy

These are attitude measures from Pew Research Center's survey of US adults, conducted on its American Trends Panel. They describe what people think, which is worth knowing precisely because it so often gets reported as what happened to them.

1. 81% of Americans are concerned about how companies use their data.

(Source: Pew Research Center)

Pew found 81% of US adults “feel very or somewhat concerned with how companies use the data they collect” about them.

“Very or somewhat” is a wide band, and combining the two produces a larger headline than either alone, which is standard practice but worth noticing. The figure replaces an older 73% concern statistic on this page that measured a different question.

2. 71% are concerned about how government uses their data.

(Source: Pew Research Center)

The same survey found 71% concerned about the government's use of data collected about them, ten points below the figure for companies.

Keeping these apart matters. Americans are more worried about commercial data use than state data use, which is the opposite of what a page framed around surveillance would lead you to expect, and the two should never be merged into a single “privacy concern” number.

3. 73% feel they have little or no control over company data.

(Source: Pew Research Center)

73% of US adults say they have very little or no control over the data companies collect about them.

Control is a distinct measure from concern, and the gap between them is the interesting part: slightly fewer people feel powerless than feel worried, which suggests a minority are concerned but believe they can do something about it.

4. 79% feel they have little or no control over government data.

(Source: Pew Research Center)

On government-collected data, 79% say they have very little or no control.

Note the inversion against stat 2. Americans are less concerned about government data use than commercial use, but feel less able to do anything about it. Concern and control move independently, which is why the old page's habit of converting both into “trust” lost the distinction.

5. 72% have little or no understanding of data privacy laws.

(Source: Pew Research Center)

Pew found 72% of Americans “have little to no understanding about the laws and regulations that are currently in place to protect their data privacy”.

This is the most consequential attitude figure on the page. Privacy regimes in the US rely heavily on individuals exercising rights, opting out, requesting deletion, filing complaints, and that model assumes a public that knows the rights exist. Nearly three quarters say they do not.

Privacy Behaviour and Actual Incidents

Attitudes are one thing; what people do, and what actually happens to them, is another. Both sets of figures below come from the same Pew survey, which makes them directly comparable with the attitudes above.

6. 56% click “agree” on privacy policies without reading them.

(Source: Pew Research Center)

Pew found 56% of Americans “always, almost always or often click ‘agree’ right away, without reading what privacy policies say”.

Set beside the 81% who say they are concerned about company data use, this is the clearest privacy-paradox finding available: the same population reports high concern and low engagement with the one mechanism offered to them. That is not necessarily inconsistency on their part, which the next stat helps explain.

7. 61% say privacy policies do not explain what companies do.

(Source: Pew Research Center)

Roughly 61% of US adults view privacy policies as not too or not at all effective at explaining how companies use personal data.

Which reframes the previous figure. If a majority consider the documents ineffective at communicating anything, not reading them looks less like apathy and more like a reasonable assessment of their value. The failure being measured is arguably the policies', not the readers'.

8. 11% had an account taken over in the past year.

(Source: Pew Research Center)

11% of US adults reported someone taking over their email or social media account without permission in the previous 12 months.

This is an incident measure, not an attitude, and it is narrower than it is usually reported as. It covers email and social accounts specifically, not bank fraud, not card fraud, not every kind of hacking. The old claim on this page that “more than 20% of online users are victims of account hacking” was both higher and vaguer than any current measure supports.

9. 34% experienced at least one hacking or fraud incident.

(Source: Pew Research Center)

Taking a wider definition, Pew found 34% had experienced at least one of several specified problems in the past year, combining credit card fraud, account takeovers and someone attempting to open a line of credit or apply for a loan in their name.

Be careful not to collapse this into stat 8. The 34% is a combined figure across several distinct incident types; the 11% is account takeover alone. Quoting 34% as an “account hacking rate” triples the real number.

Data Breaches in 2025

The Identity Theft Resource Center tracks publicly reported US data compromises and publishes an annual count. Its 2025 report produced two figures that move in opposite directions, which makes it an unusually good illustration of why these measures cannot be merged.

10. The ITRC recorded 3,322 US data compromises in 2025.

(Source: Identity Theft Resource Center, 2025 Annual Data Breach Report)

The ITRC tracked 3,322 data compromises in the US during 2025, an all-time record. That is up from 3,152 in 2024, beats the previous high of 3,202 set in 2023, and represents a 79% increase over five years.

A compromise is one incident at one organisation, whatever its size. The count says how many separate events were publicly reported, not how many people were affected, and the two can diverge sharply, as the next figure shows.

11. Those compromises produced 278.8 million victim notices, down 79%.

(Source: Identity Theft Resource Center)

The 2025 compromises generated 278,827,933 victim notices, a 79% fall from the 1,367,117,021 issued in 2024. The ITRC attributes the drop to the absence of “mega-breaches” in 2025 of the kind that dominated 2024.

So a record number of incidents produced a fraction of the previous year's notices. Both statements are true, and reporting either alone gives a misleading picture of the year. Note also that notices are not unique people: one person caught in three compromises receives three notices, so the figure always exceeds the number of individuals affected and cannot be read as “278.8 million Americans”.

12. 70% of breach notices did not say how the attack happened.

(Source: Identity Theft Resource Center)

Of the 2025 compromises, 70% (2,324) of breach notices included no information about the nature of the attack, up from 65% (2,049) in 2024.

This is the quiet finding in the report and it undercuts a great deal of breach commentary. If seven in ten notices do not say what happened, then any published breakdown of attack methods is extrapolating from the minority that did disclose, and the disclosing minority is unlikely to be representative.

13. Financial services was the most breached sector.

(Source: Identity Theft Resource Center)

By sector in 2025: financial services led with 739 compromises, followed by healthcare (534), professional services (478), manufacturing (299) and education (188).

Sector counts reflect both how often organisations are attacked and how strictly they are required to report, so heavily regulated industries appear high partly because disclosure rules make their incidents visible. Our data breach statistics cover the wider picture.

Ways to Protect Your Privacy Online

  • Use a password manager and a unique password everywhere. Reuse is what turns one breach into several, and with 3,322 compromises recorded in a single year, assuming at least one of your accounts is caught is realistic.
  • Change a password when there is a reason to, such as a breach notice or evidence of compromise. Current NIST guidance tells services not to force routine periodic changes, because scheduled resets push people into predictable variations. Our password statistics go into the standard in detail.
  • Turn on multi-factor authentication, preferring an authenticator app or security key over SMS codes, which can be captured by a convincing fake login page.
  • Use end-to-end encrypted messaging where the content matters.
  • Treat your email address and phone number as identifiers, not contact details. They are the keys used to link your records across services and to reset your accounts.
  • Limit what you publish about yourself. Birth date, birthplace and the answers to common security questions are all reusable by someone impersonating you.
  • Be sceptical before clicking, and check the address bar before entering credentials. Credential phishing depends on you not looking.
  • Keep systems and software updated. Unpatched software is now the single most common way breaches begin.
  • Know that incognito mode is not privacy. It stops your browser keeping local history; it does not hide activity from sites, employers or your network provider. A VPN moves that visibility to the VPN provider rather than removing it.

Wrap Up

Americans are broadly worried about data privacy, more about companies (81%) than government (71%), and mostly feel unable to do much about either. Nearly three quarters say they do not understand the laws meant to protect them, and a majority accept privacy policies without reading them, largely because they do not believe the policies explain anything useful.

On the incident side, 2025 set a record for data compromises at 3,322 while victim notices fell 79%, because the year lacked the mega-breaches that inflated 2024. Both facts are true at once, and either one on its own tells a misleading story about the year.

Which is the point worth carrying away from a page like this. Concern is not incidence. Notices are not people. Account takeover is not all fraud. A record number of breaches does not mean a record number of victims. Most alarming privacy statistics in circulation are accurate measures of something other than what they are being used to claim.

About Nick Galov

Author

Unaware that life beyond the internet exists, Nick is poking servers and control panels, playing with WordPress add-ons, and helping people get the hosting that suits them.

View all posts by Nick Galov →

About Paul Tsurkan

Editor

Editorial contributor at WebTribunal.

View all posts by Paul Tsurkan →