Phishers have found a clever way to spoof Google — and their emails pass all security checks.
A new DKIM replay phishing attack abuses Google’s own OAuth infrastructure to send fake messages that look 100% legitimate, including passing DKIM authentication.
What happened:
- A phishing email was sent from “no-reply@google.com”
- It appeared in the user’s inbox alongside real Google security alerts
- The message linked to a fake support portal hosted on sites[dot]google[dot]com — a Google-owned domain
- The attacker used Google OAuth to trigger a real security alert to their inbox, then forwarded it to victims
Why this matters:
- DKIM only verifies the headers, not the envelope — allowing this spoof to work
- The phishing site was nearly indistinguishable from Google’s actual login portal
- Because the message was signed by Google and hosted on a Google domain, it bypassed most users’ suspicions
- Similar tricks have been used with PayPal and other platforms, raising broader concerns
Google has since acknowledged the issue and is working on a fix. But this attack is a reminder:
Even the most secure-looking emails can be fraudulent.
Even Google-signed emails can be weaponized.
At @Efani, we advocate for layered defense — because no one layer is ever enough.