Plan this weekAccount takeover

Fake OpenAI, SpaceX and Anthropic events are stealing work-account sessions. Never type a device code a web page gives you

Push Security reports a phishing campaign that sends invitations to fake AI events (free ChatGPT Pro, Grok Heavy, a "SpaceX share") and asks you to sign in with your work Microsoft or Google account. The pages use a device-code trick and a login proxy to take your session, even with MFA on. What it looks like and how to defend.

Updated 6 Nov 2026: see what changed

How urgent

Plan this week

Who is affected

Anyone who signs in to Microsoft or Google with a work or personal account and receives an event invitation that offers a reward for signing in. Teams that allow Microsoft device-code sign-in without needing it

What to do

Do not sign in to claim an event reward, and never enter a device code that a web page shows you. If you did, sign out of every session, change the password and tell your admin. Turn off device-code sign-in where you do not use it and use security keys (FIDO2) or passkeys

Published 6 Nov 2026 · Sources last checked 6 Nov 2026No. 59 / 87

Updates, follow-ups and corrections

  1. Update · 6 Nov 2026. First version of the story, based on Push Security's analysis (read directly) and BleepingComputer's report of 5 Nov 2026. Push sells phishing-detection products, so its own defensive advice is limited to its product; the general advice below comes from BleepingComputer and from us.

We keep the original text. When something changes, when we follow a story, or when we get something wrong, we add a dated note here instead of editing silently.

What happened

Push Security says it detected and blocked visits to fake event registration pages in late October 2026. The pages pose as an OpenAI academy event, a private SpaceX briefing and an Anthropic event tied to AWS re:Invent. They offer a reward for signing in with a work account: three months of ChatGPT Pro, three months of Grok Heavy, or a "SpaceX share". The invitations arrive by email and the links use trusted redirects, such as Google short links, Zapier links and real Zoom event pages.

The visitor first sees a bot check, then buttons to continue with Microsoft or Google:

  • Microsoft: the page shows a device code and tells you to enter it on Microsoft's real login page. Doing so hands the attacker a session. You are then sent to a second fake Microsoft page that captures the password and session cookie.
  • Google: a login proxy relays Google's real page and captures the password, the MFAMFA (multi-factor authentication) A second proof that it is really you, on top of your password, such as a code from an app or a security key. Tap again to open the glossary response and the session cookie.

Push names two kits behind the pages (ZoomMock and MSDRoot) and two phishingPhishing A message or page that pretends to be someone you trust to get you to hand over a password, a code or a payment. Tap again to open the glossary services they rely on (Kali365 and VoidProxy), and says the same operator likely runs both kits. The operator switched themes between Zoom, Anthropic, OpenAI and SpaceX between 20 and 30 October and rotates domains every few weeks. Push believes there are more lures than it has seen.

Why MFA does not save you here

The attacker is not guessing your password. With the device-code page you approve the sign-in yourself on a real Microsoft page, and with the proxy the attacker sits between you and the real login and keeps the session cookie that proves you passed MFA. That is why SMS and app codes are not enough against this kind of page. Security keys and passkeysPasskey A way to log in with your fingerprint, face or device PIN instead of a password, tied to the real site so it cannot be handed to a fake one. Tap again to open the glossary are tied to the real site, so a look-alike page cannot use them.

What to do

  1. Do not sign in to claim an event reward. A real event does not need your work account to give you a prize.
  2. Never type a device code that a web page shows you into a Microsoft login page. Device codes are meant for a TV or a command line that you started yourself.
  3. If you did it: sign out of all sessions (Microsoft and Google both have a "sign out everywhere" option), change the password, remove unknown devices and apps, and tell your admin.
  4. Admins: block the device-code flow where nobody needs it (a Conditional Access policy in Microsoft Entra), and move people to security keys or passkeys.
  5. Check the address before you sign in. The pages used event-style domains, not openai.com, spacex.com or anthropic.com.
  6. Report the email and do not forward the link.

What we could not confirm

We read Push's analysis directly; it is vendor research and we could not check its findings with another source. It does not say how the first invitations reach people, or how many accounts were taken. Push says the operator likely uses services sold to many customers, so a domain or service name does not identify one group. We did not open the pages.

Comments

No account needed. Add a nickname if you like, or leave it empty to comment anonymously. Comments are reviewed before they appear. Links are not allowed.

    We store your comment, your nickname and a scrambled form of your connection that we use only to limit spam. We do not store your IP address or set cookies. Cloudflare Turnstile checks that you are not a bot.

    Sources and verification

    We use primary sources first and say when we rely on a news report. Dates are the day we last checked each source.

    Spotted a mistake? Email [email protected] and we will correct it and note the change.

    Get the free runbooks

    One-page security checklists for founders who build with AI. You get a confirmation email first, then the runbooks.

    Send me the runbooks