Company Updates

Push Security spots phishing campaign using fake AI event invites to steal Microsoft and Google logins

Share
Push Security spots phishing campaign using fake AI event invites to steal Microsoft and Google logins

Push Security detected a phishing campaign in late October 2026 that used fake AI event invitations from OpenAI, SpaceX and Anthropic to trick employees into surrendering Microsoft and Google work account access. The attackers routed victims through Google and Zapier links before presenting Cloudflare challenges and professional-looking registration pages.

TL;DR

  • Push Security documented a phishing campaign in late October 2026 that impersonated OpenAI, SpaceX and Anthropic event invites to steal Microsoft and Google work account credentials.
  • Attackers used Kali365 for Microsoft device-auth phishing and VoidProxy for session-cookie theft, bypassing multifactor authentication in some cases.
  • The campaign leveraged familiar services like Google and Zapier links, Cloudflare challenges, and branded registration pages to appear legitimate.

What happened

Push Security reported that attackers created convincing event registration pages impersonating OpenAI, SpaceX and Anthropic, which were detected and blocked in late October 2026. The pages promised rewards like three months of ChatGPT Pro or Grok Heavy to lure registrants.

Visitors were routed through shortened Google and Zapier links, then presented with Cloudflare security challenges before reaching the fraudulent registration pages. One fake OpenAI event advertised a November 4 livestream called 'ChatGPT for Social & Digital Marketing Teams,' while another promoted a November 12 SpaceX briefing with Grok Heavy or 'SpaceX share' rewards.

For Microsoft logins, the phishing site used Kali365 to generate a device authentication code that victims were instructed to enter on Microsoft's legitimate sign-in page, enabling attackers to capture authentication tokens without the password.

For Google logins, the site opened a VoidProxy pop-up that mimicked Google's sign-in screen, capturing credentials and multifactor responses to steal session cookies for persistent access.

Why it matters

This campaign shows how attackers are weaponizing AI event hype to bypass multifactor authentication and gain persistent access to corporate accounts, putting sensitive emails, documents and connected apps at risk.

Developers and startup founders who frequently attend AI webinars should treat unexpected event invites with skepticism, especially when they request corporate account sign-ins or offer premium AI subscriptions as rewards.

Investors evaluating AI startups should be aware that credential theft via phishing remains a top risk vector, and companies must enforce phishing-resistant authentication and continuous monitoring to mitigate such attacks.

The use of Kali365 and VoidProxy highlights the commoditization of phishing infrastructure, making it easier for criminals to launch sophisticated campaigns without deep technical skills.

Key facts

  • Push Security detected and blocked visits to malicious event registration pages in late October 2026, according to the company's report.
  • The fake OpenAI event promised three months of ChatGPT Pro to participants in a November 4 livestream titled 'ChatGPT for Social & Digital Marketing Teams'.
  • The fake SpaceX event offered three months of Grok Heavy or a 'SpaceX share' to registrants for a November 12 briefing.
  • Attackers used Kali365, a phishing-as-a-service platform, to generate Microsoft device authentication codes for victims to enter on Microsoft's legitimate sign-in page.
  • VoidProxy was used to capture Google credentials, multifactor responses, and session cookies via a fake Google sign-in pop-up.
  • The campaign routed visitors through shortened links associated with Google and Zapier, and presented Cloudflare security challenges on the fraudulent registration pages.
  • Push Security identified two phishing kits, ZoomMock and MSDRoot, used to create the fake event pages, with evidence suggesting the same operators were behind both kits.
  • Microsoft told reporters it takes such threats seriously and that Microsoft Defender includes detections to block the identified threats.

Context

AI event invitations have become a prime lure for phishing campaigns as interest in generative AI tools surges among enterprise users, making employees more likely to click on unexpected invites.

Credential theft via sophisticated phishing remains one of the top vectors for account takeovers despite widespread adoption of multifactor authentication, as attackers adapt to bypass newer defenses.

The commoditization of phishing infrastructure, exemplified by services like Kali365 and VoidProxy, lowers the barrier to entry for cybercriminals and increases the volume of targeted attacks.

Topics

Join the discussion

Have a take on this story? Weigh in with our community on Facebook.

💬 Discuss on Facebook →