Fake OpenAI and SpaceX Event Invites Used in Phishing Campaign to Steal Microsoft and Google Accounts

Hackers Deploy Sophisticated Phishing Campaign Using Fake AI Event Invitations

A new phishing campaign is targeting employees by exploiting their interest in artificial intelligence tools through fake event invitations from OpenAI, SpaceX, and Anthropic. Security researchers at Push Security detected and blocked visits to malicious event registration pages in late October 2024, uncovering a sophisticated operation that uses two separate phishing-as-a-service platforms to steal corporate account credentials1

1

The attackers create convincing registration pages for nonexistent events, including an OpenAI Academy livestream titled "ChatGPT for Social & Digital Marketing Teams," a private SpaceX briefing scheduled for November 12, and an Anthropic presentation at AWS re:Invent 2024. These fake event invitations lure victims with enticing rewards: three months of ChatGPT Pro, access to Grok Heavy, or even SpaceX shares2

2

Dual Phishing-as-a-Service Platforms Target Microsoft and Google Accounts

What distinguishes this phishing campaign is its use of two custom phishing kits, tracked as ZoomMock and MSDRoot, which integrate separate phishing-as-a-service platforms. Kali365 targets Microsoft accounts through device code phishing attacks, while VoidProxy operates as an adversary-in-the-middle platform to intercept Google sign-ins and steal corporate account credentials1

1

Researchers believe the same threat actors operate both kits, as they use shared infrastructure and identical targeting methods. The fake registration pages tell visitors they must sign in with their work accounts to register and claim the advertised rewards, presenting two options: "Continue with Microsoft" and "Continue with Google"2

2

Device Code Phishing Exploits Microsoft Authentication Flow

When victims select the Microsoft option, the phishing campaign initiates a device code phishing attack through Kali365. This technique abuses the legitimate OAuth 2.0 device authorization grant flow to obtain authentication tokens for the victim's Microsoft Entra account. The visitor receives a device code and instructions to enter it on Microsoft's legitimate device login website, where they authenticate and approve the request1

1

This allows the Kali365 platform to steal Microsoft authentication tokens without requiring victims to disclose their password directly. After the device code authentication succeeds, the page redirects the victim to a VoidProxy Microsoft login page, where attackers capture the victim's password and session cookies. The stolen tokens can be used to access Microsoft 365, Salesforce, Google Workspace, Dropbox, Adobe, SAP, Slack, Zendesk, Atlassian, and many other connected SSO applications1

1

Adversary-in-the-Middle Attacks Bypass Multi-Factor Authentication on Google

Victims who choose "Continue with Google" encounter a different attack vector. The phishing campaign opens a pop-up containing a Google login page from VoidProxy, which operates as an adversary-in-the-middle platform that sits between the victim's browser and Google's login page. VoidProxy proxies the login forms and captures credentials, MFA responses, and session cookies, effectively allowing attackers to bypass multi-factor authentication1

1

The platform relays the authentication process while capturing the victim's password, multi-factor authentication response, and session cookies, which allows the attackers to obtain an authenticated session without needing to move past MFA. Both the Kali365 and VoidProxy platforms enable attackers to maintain persistent access to compromised accounts2

2

Related Stories

OpenAI Impersonation Phishing Attack Targets Businesses Globally

Fake ChatGPT, Gemini Sites Steal Advertising Accounts via Browser-in-Browser Attacks

AI-Powered Gamma Platform Exploited in Sophisticated Phishing Attack Targeting Microsoft SharePoint Users

Chinese Hackers Impersonate Anthropic Exec and White House Official to Target AI Policy Experts

AI-Themed Lures Exploit Employee Interest in Emerging Technologies

The attackers distribute phishing links through trusted services, including Google and Zapier URL shorteners, to avoid detection by security scanners and URL reputation systems. Before displaying the fake registration pages, the websites show a Cloudflare CAPTCHA to "Confirm you're not a robot," preventing automated security platforms from scanning the phishing site1

1

"Employees want access to the latest AI tools, making them excellent bait for attacks," Push Security stated in its analysis. The campaign represents an evolution of AI-themed lures that have been increasingly exploited since ChatGPT's release five years ago. In July 2024, ANY.RUN documented a similar campaign where attackers abused legitimate Zoom event pages to promote fake Meta, OpenAI, and Anthropic events. More recently, Cofense reported on fake event invitations impersonating Anthropic that were distributed through GoTo Webinar's email infrastructure2

2

Organizations Face Growing Threat from MFA-Bypassing Techniques

Push Security warns that these detection evasion techniques, combined with MFA-bypassing adversary-in-the-middle and device code phishing payloads, mean this attack is designed to circumvent the controls that most organizations rely on to stop phishing attacks today. Device code phishing attacks have been used in the past for data theft and cyberespionage attacks, and the combination with adversary-in-the-middle techniques represents a significant escalation in threat sophistication2

2

As AI tools penetrate deeper into every aspect of private and business lives, security experts anticipate that threat actors will increasingly abuse these platforms in future campaigns. Organizations should watch for suspicious event invitations, verify the authenticity of AI-related offers through official channels, and implement additional security controls beyond traditional multi-factor authentication to protect against session cookie theft and token-based attacks.