MFA fatigue: push notification bombing
An attacker who knows your password floods your phone with approval requests until you accept. How to recognize it and what to do.
On this page
In 30 seconds
- Where it happens
- On your phone or in your authentication app, as login notifications you did not request[1].
- What the attacker wants
- A single approval. Once the request is approved, the attacker reaches the targeted resource[1].
- If the notifications keep coming
- Report it right away rather than approving one[1].
What it is
An attacker who already has an employee's password runs into the phone approval step. MFA fatigue gets around it without breaking it: the attacker triggers approval requests in a burst until the person finally accepts to make them stop[1]. The technique is also called MFA bombing, push spam or prompt bombing[1].
MITRE ATT&CK lists it under the code T1621. It abuses the automatic sending of notifications by services such as Duo Push, Microsoft Authenticator or Okta, so that the user grants access to their own account[2].
Straight answers
What is MFA fatigue?
It is a form of social engineering. The attacker uses persistence and saturation to disrupt the user's ability to log in[1]. The authentication system works normally. What the attacker targets is the person who decides, not a software flaw[1].
How does the attacker get the password?
According to Proofpoint, the attacker may obtain it through brute-force attacks, targeted phishing or by buying stolen credentials on the dark web[3]. MITRE adds a case where the attacker has no credentials: the attacker can also abuse notification sending when that option is enabled for self-service password reset[2].
Is a single approval really enough?
Yes. Once a single message is approved, the attacker reaches the targeted resource[1]. As long as the attacker keeps an authenticated session open, they can move around the system, steal information, change settings and add access without any further approval request[1].
What should I do when the notifications arrive?
Do not approve to make them stop. MITRE recommends training users to accept only requests that come from logins they started themselves, to check where the attempt comes from, and to report suspicious or unsolicited requests[2]. Proofpoint says the same: reject unexpected requests and report anything strange immediately, instead of approving one to stop the flood[1].
Are all phone approvals equal?
No. CISA ranks forms of multifactor authentication from strongest to weakest. A push notification with no number to enter is vulnerable to push bombing, while a push notification with number matching resists it[4]. In the second case, the user must enter in the app digits displayed by the identity platform[4]. FIDO/WebAuthn keys and public key infrastructure (PKI) authentication are the phishing-resistant forms[4].
Do you have to be careless to fall for it?
No. According to Proofpoint, even experienced and trained users can be tired, distracted or on autopilot, and a single reflex tap on "approve" is enough[1].
Inside the attacker's head
The attacker starts from valid credentials: phishing, password reuse or stolen lists bought online[1]. The attacker then tries to log in to a cloud application, a VPN or a single sign-on portal. The platform does what it is supposed to do and sends a notification to the user. All the attacker has to do is repeat the login[1].
The technical effort is low. Proofpoint notes that it is relatively easy to write a script that keeps trying to log in and makes enough noise to wear the victim down[1]. The same page cites a Proofpoint study: almost half of the hijacked accounts had multifactor authentication enabled, while 89% of security professionals consider it comprehensive against account takeover[1].
The attacker can also add a voice. Proofpoint describes cases where the attacker poses as IT support and asks the victim to approve the request to fix a problem or avoid an account lockout[1]. This impersonation of support is a known method of Scattered Spider: according to public information reported by CISA, the group posed as IT or help desk staff by phone or SMS to obtain credentials, or to convince employees to share their one-time passcode[5].
What you see on screen
Proofpoint describes the following signs[1]:
- several notifications in a short time, although you are not trying to log in;
- requests at unusual hours, for example at 2 a.m. or during time off;
- requests that come back every few seconds or minutes, sometimes for hours;
- an "Approve / Deny" request with no detail about the device, the location or the IP address, or with no number to enter;
- notifications you cannot manage to stop;
- a message or a call claiming to come from support and asking you to approve a pending request.
MITRE specifies that the attacker may also flood the victim with SMS messages and phone calls[2]. According to Proofpoint, a person who is very busy and receives several notifications in a row is even more likely to accept without examining the request[3].
Under the hoodIT teams
MFA fatigue does not break two-factor authentication. Proofpoint sums it up: the system works as intended, but the problem is the person's behavior under pressure[1]. In many organizations, the notification is the only second check, so the single "approve or deny" decision becomes the weak point[1].
MITRE classifies technique T1621 under credential access. It recommends replacing one-click approvals with safer mechanisms, for example entering a code shown on the login screen, changing default configurations and limiting the number of requests sent to a user over a given period[2]. It also suggests conditional access rules that block logins from suspicious locations or non-compliant devices[2]. Proofpoint suggests restricting the number of requests per hour for an account to between 1 and 5, beyond which the account may be locked for a time, the security team alerted or extra authentication required[1].
On the defense side, CISA recommends adopting phishing-resistant multifactor authentication; FIDO/WebAuthn keys and PKI authentication are not subject to push bombing[5]. Once inside an account, Scattered Spider registers its own multifactor authentication tokens to establish a lasting foothold[5].
Case filesReal cases
Uber, September 2022
According to Proofpoint, which reports Uber's own statement, the company said in September 2022 that an attacker linked to the Lapsus$ group had compromised a contractor's account[3]. Uber said the attacker had possibly bought the credentials on the dark web[3]. The contractor received several notifications while the attacker tried to log in, and eventually accepted one[3]. Once logged in, the attacker reached other accounts and obtained higher privileges[3].
Cisco, May 2022
Proofpoint reports that Cisco suffered an intrusion in May 2022 that traced back to an employee's compromised Google account[3]. The target received a large number of voice phishing attempts and notifications, and gave in to MFA fatigue[3]. According to Proofpoint, Cisco said the attackers stole non-sensitive data[3].
Microsoft, March 2022
According to Proofpoint, Microsoft confirmed in March 2022 that Lapsus$ had breached its network and stolen proprietary source code[3]. The intrusion traced back to a user whose account had been compromised and who then suffered an MFA fatigue attack[3].
Scattered Spider, joint advisory of November 2023
CISA published on 16 November 2023, with the FBI and partner agencies, an advisory on Scattered Spider, a criminal group that targets large companies and their outsourced IT help desks[5]. Among its methods, the advisory lists sending authentication notifications repeatedly until the employee presses "Accept"[5]. The group also convinced mobile carriers to transfer a victim's number to a SIM card it controlled[5]. The advisory was updated through 29 July 2025[5].
Sources
- Proofpoint, “What Is an MFA Fatigue Attack?”. https://www.proofpoint.com/us/threat-reference/mfa-fatigue-attack (accessed 14 November 2026)
- MITRE ATT&CK, “Multi-Factor Authentication Request Generation”. https://attack.mitre.org/techniques/T1621/ (accessed 14 November 2026)
- Proofpoint, “Preventing MFA Fatigue Attacks: Safeguarding Your Organization”, 21 November 2023. https://www.proofpoint.com/us/blog/information-protection/preventing-mfa-fatigue-attacks (accessed 14 November 2026)
- CISA, “Implementing Phishing-Resistant MFA”. https://www.cisa.gov/sites/default/files/publications/fact-sheet-implementing-phishing-resistant-mfa-508c.pdf (accessed 14 November 2026)
- CISA, “Scattered Spider”, 16 November 2023. https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a (accessed 14 November 2026)
Cite this article
Brain, “MFA fatigue: push notification bombing”, Cyberpedia, 2026. https:/
BibTeX
@misc{brain_cp005_en_2026,
author = {Brain},
title = {MFA fatigue: push notification bombing},
year = {2026},
howpublished = {Cyberpedia},
url = {https://startbrain.io/cyberpedia/en/articles/mfa-fatigue-push-bombing/},
note = {CC BY 4.0}
}