Hackers are abusing legitimate email systems to deliver phishing content through order confirmations, account notifications, and delivery failure reports.
Instead of relying on fake sender addresses, attackers place malicious instructions inside messages generated by trusted services.
Cisco Talos tracked abuse involving more than 25 legitimate platforms over the past year, according to the supplied research. These included collaboration, productivity, file sharing, survey, payment, and artificial intelligence services.
The activity expands a technique Talos calls platform-as-a-proxy, where attackers use trusted infrastructure to carry their content.
Earlier research documented similar abuse of GitHub and Jira notification systems, allowing malicious messages to arrive through legitimate delivery channels.
Hackers Weaponize Email Notifications
Talos examined two campaigns abusing Staples and Dell order notifications, together accounting for roughly 1,400 observed messages.
The attackers inserted phone numbers and support instructions into fields normally reserved for shipping or billing information.
One Staples message appeared to confirm an Apple iPad purchase. It included realistic pricing, delivery details, and company branding, while displaying “Helpline 806 283 7712” within the order information.
A Dell notification presented an unexpected Microsoft Project Standard 2024 purchase totaling $736.09. Its shipping and billing sections contained “Helpline 808 232 6397” and instructions to contact a sales team.
Both campaigns aimed to make recipients call an attacker-controlled number rather than click a phishing link. The unexpected purchase created urgency, while the familiar retailer template made the supplied contact details appear credible.
A resulting call could expose victims to credential theft, payment scams, or fraudulent remote assistance. However, the supplied findings describe these as possible follow-on activities, not confirmed outcomes.

Authentication checks reinforced the messages’ apparent legitimacy. The samples showed passing DKIM and DMARC results associated with retailer infrastructure. These checks supported the sender’s authenticity but did not establish whether the embedded phone numbers were safe.
Attackers also repurposed list bombing, which floods a target’s inbox through registration and subscription forms. By placing scam text inside name, company, or project fields, they caused legitimate platforms to repeat that content in automated emails.
This turns a notification workflow into a delivery channel. The service supplies its reputation and email infrastructure, while the attacker supplies the lure.
Delivery failure reports create another opportunity. These messages, also called delivery status notifications or non-delivery reports, may include original headers, message text, attachments, or an embedded copy of an undelivered email, talosintelligence said.
Talos observed a delivery notification containing a WeTransfer-themed lure. Potential delivery mechanisms include backscatter sent to a forged envelope sender and deliberately triggered delivery failures. The supplied findings do not establish which mechanism produced that example.
Defenders should separate sender trust from content trust. Platform operators should validate user-controlled fields before storing them and before inserting them into email templates.
Join 16,000+ SOC teams using ANY.RUN to streamline threat investigations and reduce manual effort. Explore for your team