# MFA Prompt Bombing: How Attackers Turn Your Second Factor Against You


Multi-factor authentication (MFA) was supposed to be the great equalizer in identity security. Even if attackers stole your password—which, let's be honest, many of them have—they still couldn't access your accounts without that critical second factor. The logic was sound. The execution, however, left a dangerous gap that attackers have been systematically exploiting.


Today, prompt bombing attacks represent a live threat to organizations of all sizes. Unlike traditional credential theft, this technique doesn't require attackers to intercept or forge authentication codes. Instead, it weaponizes the most vulnerable part of any security system: human decision-making under pressure.


## The Threat: Understanding Prompt Bombing


Prompt bombing is elegantly simple in its execution. An attacker repeatedly triggers multi-factor authentication (MFA) push notifications to a target's phone, attempting to wear down the user's resistance or create confusion about the legitimacy of the request. The goal is straightforward: get the target to approve a login attempt they didn't initiate.


The attack works because push-based MFA—the kind that sends a "tap to approve" notification to your phone—lacks critical contextual information. When a user receives a prompt, they typically see little more than a timestamp and a generic message. There's no clear indication of:


  • Where the login is coming from (geographic location, IP address, device type)
  • What application is being accessed
  • Whether the request was initiated by them or an attacker
  • Whether previous prompts were legitimate or malicious

  • When a user receives ten notifications in rapid succession, the natural human instinct is to assume something is malfunctioning. When that confusion is combined with a well-timed phone call from someone claiming to be IT support, the scenario becomes genuinely difficult to assess—especially for employees who aren't security specialists.


    ## Background and Context: The MFA Promise vs. Reality


    MFA adoption has skyrocketed over the past decade, and for good reason. The widespread use of stolen credentials from breaches means that passwords alone are no longer a viable security control. Organizations across every sector have implemented MFA as a foundational security layer.


    The problem is that not all MFA implementations are created equal.


    Push-based MFA (used by platforms like Microsoft Entra ID, Okta, Duo, and many VPN providers) was designed for user convenience. A notification arrives on your phone, you tap "approve," and you're logged in. No codes to enter, no apps to open. It's frictionless.


    Phishing-resistant MFA, by contrast, requires either:

  • FIDO2 security keys (hardware devices like YubiKey or Google Titan)
  • Hardware tokens that generate time-based codes
  • Number-matching codes from authenticator apps that require the user to manually verify a unique number displayed on the login screen

  • These approaches are significantly harder to abuse because they either require physical hardware the attacker doesn't possess or involve explicit code verification that creates additional friction and context for the user.


    The industry's rush toward push-based MFA created a false sense of security. The attack vector was always there; it just took time for organized threat actors to refine and systematize the technique.


    ## Technical Details: How the Attack Works


    Prompt bombing requires three critical elements:


    1. Valid Account Credentials

    The attacker needs a real username and password. These typically come from:

  • Password dumps stolen from previous breaches (widely available on the dark web)
  • Credential harvesting through phishing campaigns
  • Compromised personal accounts syncing stored credentials (as we'll see in the Cisco case)

  • 2. A Vulnerable Login Portal

    The target organization must use push-based MFA on an accessible service such as:

  • VPN gateways
  • Microsoft 365 or other cloud services
  • Identity management platforms (Okta, Azure AD)
  • Remote desktop protocols (RDP) with push MFA

  • 3. A Target Who Can Be Manipulated

    The attacker attempts to overcome user resistance through:

  • Prompt fatigue: Sending multiple notifications in rapid succession, hoping the user approves one by accident or assumes it's a system malfunction
  • Vishing (voice phishing): Calling the target while bombing prompts, claiming to be IT support and saying something like "I'm helping you reset your password, just accept that notification when it comes"
  • Social engineering: Creating a convincing scenario where the user's natural inclination is to approve the request

  • Once the user approves a single prompt, the attacker is logged in as that user. From that point forward, the account is compromised—and critically, legitimate security systems typically won't flag this as suspicious because the login looks entirely normal.


    ## Case Study: The 2022 Cisco Breach


    The most instructive example of prompt bombing's effectiveness comes from the 2022 Cisco breach, which exposed the technique's lethality against even mature security programs.


    How the Attack Unfolded:


    An attacker affiliated with the Yanluowang ransomware group compromised a Cisco employee's personal Google account. The employee had configured their browser to sync passwords, including the credentials for their Cisco VPN. Armed with these valid credentials, the attacker attempted to log in to Cisco's VPN and began sending push MFA prompts to the employee's phone.


    The initial prompts were rejected. Rather than giving up, the attacker escalated to vishing calls. They posed as representatives from trusted support organizations, often speaking in different accents to seem more credible, and eventually convinced the employee to approve a pending push notification.


    That single approval gave the attacker VPN access as the employee.


    Post-Breach Escalation:


    Once inside the network, the attacker didn't stop. They:

  • Enrolled their own devices for MFA to maintain persistent access
  • Escalated privileges to gain administrative access
  • Accessed Citrix servers and domain controllers
  • Exfiltrated approximately 2.8GB of sensitive data before being detected and evicted

  • The Cisco breach is significant not because Cisco lacked security controls—the company is hardly known for weak security posture—but precisely because it demonstrates that prompt bombing can work even against well-resourced organizations with mature security programs. If it worked against Cisco, it can work against you.


    ## Implications for Organizations


    Widespread Exposure


    Any organization using push-based MFA for remote access, cloud services, or privileged account management is vulnerable. This includes:

  • Technology companies
  • Financial institutions
  • Healthcare organizations
  • Government agencies
  • Educational institutions

  • Essentially, if your organization supports remote work or cloud infrastructure, you're in scope.


    The Convergence of Attack Vectors


    Prompt bombing is rarely deployed in isolation. Threat actors combine it with:

  • Leaked password databases that make credential acquisition trivial
  • OSINT (open-source intelligence) to identify and research targets
  • Vishing campaigns that increase approval rates significantly
  • Persistence mechanisms that allow attackers to maintain access after initial compromise

  • This convergence makes the attack more effective and more dangerous than any single component would suggest.


    Detection Gaps


    Modern security systems are designed to detect anomalous login patterns. However, an approved MFA prompt looks identical to a legitimate login. The attacker's traffic enters the network through a valid, approved authentication event. Security tools have no way to distinguish this from normal business activity.


    ## Recommendations: Preventing Prompt Bombing


    Organizations should implement a layered defense that addresses prompt bombing at multiple levels:


    ### 1. Migrate to Phishing-Resistant MFA


    This is the most critical recommendation. Replace push-based MFA with:

  • FIDO2 security keys for high-risk users (executives, IT staff, developers)
  • Hardware tokens for VPN and remote access
  • Number-matching codes (where the user must verify a unique number displayed both on the login screen and in their authenticator app)

  • These approaches are significantly harder to social engineer or brute-force.


    ### 2. Implement Passwordless Authentication


    For high-security access points, consider eliminating passwords entirely in favor of:

  • Windows Hello for Business
  • Conditional access policies that require additional verification based on login context
  • Biometric authentication

  • ### 3. Strengthen Identity and Access Management (IAM)


  • Deploy anomaly detection that flags unusual login patterns (new locations, devices, times)
  • Implement conditional access policies that require additional verification based on risk scoring
  • Enforce device compliance requirements before allowing access
  • Use geo-blocking where appropriate to restrict access from unusual locations

  • ### 4. Block Compromised Passwords


  • Use password breach databases to prevent users from setting passwords that appear in known breaches
  • Implement password spraying detection to identify when an account is being targeted with multiple failed attempts
  • Monitor for credential stuffing patterns and alert users when their accounts are accessed with compromised passwords

  • ### 5. Employee Security Awareness Training


  • Train employees to never approve MFA prompts they didn't initiate
  • Teach them that legitimate IT support will never ask them to approve a login they didn't just attempt
  • Create a clear process for reporting suspicious activity and ensure employees know they won't be punished for false positives

  • ### 6. Enhanced Monitoring and Response


  • Monitor for prompt bombing patterns (multiple rejections followed by an approval)
  • Establish rapid response procedures for suspicious login activity
  • Revoke sessions and force re-authentication if suspicious activity is detected
  • Require manual approval for privilege escalation attempts, even from seemingly legitimate accounts

  • ## HackWire Analysis


    The critical lesson from prompt bombing is this: security theater that relies on user judgment under pressure is not security at all. MFA was sold as the solution to credential compromise, but push-based MFA simply shifted the attack vector from "steal the code" to "convince the user to give it to you."


    The timing matters here. Prompt bombing wasn't a theoretical risk in 2022 when Cisco was breached—it was an active technique deployed by organized threat groups. Yet four years later, countless organizations are still running exclusively on push-based MFA for their most sensitive access points. The gap between threat reality and organizational implementation has widened, not narrowed.


    What's particularly insidious about this attack is that it exploits the very factors that make good employees vulnerable: trust, helpfulness, and the instinct to respond quickly when they believe IT is calling with a legitimate request. This isn't a flaw in employee judgment; it's a flaw in the authentication architecture itself.


    The path forward requires moving away from authentication systems that depend on human decision-making under pressure and toward systems that provide verifiable proof of identity. FIDO2 keys, hardware tokens, and number-matching MFA are not convenient—but they work. In security, "works" beats "convenient" every single time.


    Organizations that continue to rely exclusively on push-based MFA are making a calculated bet that prompt bombing won't target their users. That's not a security strategy; it's gambling. — *HackWire Editorial*


    ## Related Coverage


  • Read more in our [Breaches](https://www.hackwire.news/category/breaches) coverage
  • Cross-reference with [Vulnerabilities](https://www.hackwire.news/category/vulnerabilities) and [Malware](https://www.hackwire.news/category/malware)
  • Stay current via the [HackWire homepage](https://www.hackwire.news/)