A data breach is not an abstract idea: it affects real people, often for years. Here are seven Canadian cases, chosen because regulators (privacy commissioners, courts) or the organizations themselves published the facts. For each: what happened, what was found, what we take from it and what you can do.
The seven cases at a glance
Summary of the cases: year, type of incident, people affected and what the authorities concluded.
Detection and containment found too slow (special report, 2024)
Desjardins: the employee who copied the data
When: 2017 to 2019; announced in June 2019
Who: Close to 9.7 million people, according to the Commissioner
Type: Insider threat
What happened
An employee of the financial co-operative copied, over several years, member information he legitimately had access to, then passed it to a third party: names, birth dates, addresses, social insurance numbers, emails, phone numbers and transaction habits. Desjardins first spoke of more than 2.9 million members, then of 4.2 million members with an active account; the Commissioner put the total at close to 9.7 million people, because it also counts the inactive files the employee had access to.
What the authorities concluded
A joint investigation by the Office of the Privacy Commissioner of Canada and Quebec's Commission d'accès à l'information (findings of December 14, 2020) attributed the breach to a series of gaps in administrative and technological safeguards, and found shortcomings on accountability, retention periods and security safeguards. A class-action settlement approved in 2022 provides up to about $200.9M in individual claims (up to $90 per person, up to $1,000 for identity theft) and five years of credit monitoring.
What we take from it
Insider threats are real. Legitimate access can become the way out for data.
Least privilege. Everyone should access only what the job needs, and every bulk access should be detected.
Keep less. Inactive files kept too long increased the number of people affected.
A social insurance number does not change. Once exposed, the risk lasts for years.
What you can do
Place a fraud alert with Equifax and TransUnion, and monitor your credit reports.
Be careful with messages that quote your details: they can fuel targeted scams.
If you were a member, check the services offered (credit monitoring) and claim what you are entitled to.
Who: About 15 million customers, mostly in Ontario and British Columbia
Type: Ransomware
What happened
LifeLabs, a major provider of laboratory tests, detected a cyberattack on October 28, 2019. The attackers demanded a ransom and claimed to have stolen records. According to the company, the data involved included names, addresses, emails, account logins and passwords, birth dates and health card numbers, and the test results of about 85,000 Ontario customers (2016 or earlier). LifeLabs confirmed it paid a ransom, without disclosing the amount.
What the authorities concluded
In their joint report of June 25, 2020, the Ontario and British Columbia commissioners concluded that LifeLabs had not taken reasonable steps to protect personal information and personal health information, contravening the laws of both provinces, while acknowledging that the company took reasonable steps to contain the incident and investigate. They ordered improvements to its practices, and the B.C. commissioner noted that regulators should be able to consider financial penalties.
What we take from it
Health data is among the most sensitive. It can reveal very private conditions and cannot be “revoked”.
Paying a ransom guarantees nothing. There is no proof the data is really destroyed, and it funds crime.
Time matters. About seven weeks passed between detecting the attack and the public announcement.
Account passwords were in the leak: any account reused elsewhere becomes a risk.
What you can do
If you had a LifeLabs account, change that password and the password of any account where you reused it.
Be wary of calls or emails claiming to come from a clinic or lab: your health data can make them credible.
Ask your health providers how they protect your information, and use your right of access.
Who: About 6 million people in Canada, about 100 million in the United States
Type: Misconfiguration (cloud)
What happened
A former engineer of a cloud services provider exploited a faulty firewall configuration to reach credit card application data stored in Capital One's cloud. According to the company, about 6 million people in Canada and about 100 million in the United States were affected, including about one million Canadian social insurance numbers. The breach was discovered when the person posted the exploit online. Capital One offered two years of credit monitoring and identity theft insurance.
What the authorities concluded
The Office of the Privacy Commissioner of Canada opened an investigation after Capital One reported the breach and customers complained. In the United States, the banking regulator fined the bank US$80M in 2020 for risk-management failures linked to its move to the cloud.
What we take from it
Responsibility is shared. The cloud provider secures the infrastructure; the customer must configure it properly.
The basics matter. A well-configured firewall, limited access and monitoring beat a sophisticated tool.
Application data kept for years (2005 to 2019) enlarges the exposure.
What you can do
Turn on the credit monitoring offered and place a fraud alert.
For security questions, use fictional answers stored in a password manager.
Check whether your email addresses appear in known breaches (Have I Been Pwned).
Facebook and Cambridge Analytica: consent that was not consent
When: 2014-2015; revealed in March 2018; litigation ongoing
Who: About 87 million people worldwide, including about 622,000 in Canada (Facebook's figures)
Type: Sharing with third parties
What happened
A quiz app, “thisisyourdigitallife”, collected the data of a few hundred thousand Facebook users and, because of how the platform worked at the time, that of their friends. The data was passed to the firm Cambridge Analytica, which used it for political profiling. Facebook put the total at about 87 million people, including about 622,000 in Canada. Cambridge Analytica ceased operations in 2018.
What the authorities concluded
The Office of the Privacy Commissioner of Canada and its British Columbia counterpart concluded in 2019 that Facebook had broken the law. Facebook disputed the findings and refused to apply the recommendations; the Commissioner went to the Federal Court, which sided with Facebook in 2023. The Federal Court of Appeal overturned that judgment on September 9, 2024: consent was not valid and Facebook had not adequately supervised third-party apps. The Supreme Court of Canada agreed to hear the appeal, which it heard on March 19, 2026; at the time of the last check, judgment was reserved. In the United States, Facebook agreed to pay US$5 billion to the FTC in 2019; in the United Kingdom, the authority imposed a £500,000 fine.
What we take from it
Consent must be real. Accepting long terms of use that nobody reads is not informed consent.
Your privacy also depends on others. Your friends' data can reveal things about you, and the reverse.
Platforms must supervise their partners. Giving access to thousands of apps without checking them exposes everyone.
What you can do
Review the apps and sites connected to your account and remove those you no longer use.
Limit the audience of your posts and turn off tracking of your activity outside the platform.
Do not sign in to a third-party service “with Facebook” unless you need to.
Clearview AI scraped billions of face images from the internet, without people knowing, to build a facial recognition database sold to clients, including Canadian police forces. Anyone whose photo appeared somewhere online could end up in the database.
What the authorities concluded
The Office of the Privacy Commissioner of Canada, Quebec's Commission d'accès à l'information and the British Columbia and Alberta commissioners concluded, in a joint investigation (February 2, 2021), that this collection amounted to mass surveillance and broke federal and provincial laws, for lack of consent. According to their findings, at least 48 law enforcement agencies in the country had used the tool. Clearview AI disputed the findings; it had withdrawn from the Canadian market before they were published.
What we take from it
The face is biometric data. Unlike a password, you cannot change it.
“Public” does not mean “free for the taking”. A photo visible online does not authorize mass collection for other purposes.
Police use requires a clear legal framework, not discreet trials of commercial products.
What you can do
Think before posting photos of faces (yours, your children's) on public profiles.
Set your accounts to private and remove photos you no longer want circulating.
Learn more about your rights regarding biometric data.
The Tim Hortons app: location tracked in the background
When: Findings of June 1, 2022
Who: App users who had granted access to their location
Type: Geolocation
What happened
The chain's mobile app asked for access to users' location, but many believed it used it only while the app was open. In fact it collected location continuously, even in the background, as long as the device was on. The National Post reporter whose data served as an example was located more than 2,700 times in less than five months.
What the authorities concluded
The federal, Alberta, British Columbia and Quebec commissioners concluded in June 2022 that the collection broke the laws: consent was not valid, the amount of data exceeded what was necessary and the company had no robust privacy management program for the app. Tim Hortons had stopped collecting location in August 2020 and removed the technology. The commissioners asked it to delete the remaining data and put a privacy program in place.
What we take from it
“Allow location” does not say when. The “always” setting opens permanent tracking.
Location data reveals a life. Home, work, habits: they can be inferred from movements.
Collect little. A coffee-ordering app does not need to know where you go all the time.
What you can do
In your phone settings, choose “while using the app” or “never” for location.
Review each app's permissions and remove those that cannot be justified.
When: August 2020; special report to Parliament on February 15, 2024
Who: About 5,500 CRA accounts targeted, according to the government
Type: Credential stuffing
What happened
In early August 2020, attackers used username and password pairs stolen in other breaches (“credential stuffing”) to get into accounts of the GCKey authentication service and the Canada Revenue Agency. According to the government, about 5,500 CRA accounts were targeted. Hijacked accounts were used to submit fraudulent emergency benefit applications.
What the authorities concluded
The Commissioner opened two investigations in October 2020. In a special report to Parliament on February 15, 2024, it concluded that the departments had not taken the necessary steps to detect and contain the breach promptly.
What we take from it
Reusing a password is the weak link. A leak on one site can open an account on another.
Two-factor authentication stops most of these attacks. Without it, a stolen password is enough.
Public services are targets. A tax or benefits account is a door to money and identity.
What you can do
Use a unique password for every account, especially government and bank services.
Turn on two-factor authentication, or a passkey, wherever possible.
If an account was hijacked, report it to the service concerned and immediately change related passwords.
This page is informational and is not legal advice. Figures come from the sources cited under each case and may change. Last checked: November 9, 2026.