Loading story…
Threat actors are abusing legitimate email and collaboration workflows to make social-engineering lures appear trustworthy. A reported campaign weaponizes OpenAI’s “invite your team” feature by placing malicious URLs or scam phone numbers in the organization name field, causing invitations to be sent from legitimate OpenAI email infrastructure and increasing the likelihood recipients will click links or call attacker-controlled numbers; the approach is particularly effective against businesses because a single invite workflow can target multiple employees.
Separately, consumers are being targeted by a seasonal surge of tax-refund/back-tax impersonation scams delivered via SMS and email that direct victims to malicious links or fraudulent call centers to “verify” identity—harvesting SSNs and banking details for identity theft and refund fraud. Compounding the broader email-risk environment, Gmail’s spam filtering and tab categorization reportedly malfunctioned at large scale, pushing promotional mail into Primary inboxes and flagging legitimate senders with warnings; such misclassification can increase exposure to phishing by reducing the effectiveness of inbox triage and user trust signals during active scam waves.
Get the infrastructure and lures behind it.
5 events from the most recent confirmed update back to the earliest known activity.
By 2026-01-26, researchers reported a scam campaign exploiting OpenAI's team invitation workflow to send phishing-style emails from legitimate OpenAI infrastructure, using malicious links or scam phone numbers embedded in organization names.
By 2026-01-26, tax authorities and the FTC were warning consumers about a surge in tax-season scams using texts and emails impersonating the IRS and state tax agencies to steal personal and financial information.
Google later announced the Gmail issue was fixed at 9:55 a.m. Pacific time on 2026-01-24, while warning that residual effects could continue as systems recovered.
Google acknowledged the Gmail filtering problem on its official status dashboard on 2026-01-24 and said it was actively working to resolve the incident.
On 2026-01-24, Gmail's spam filtering and inbox categorization systems reportedly malfunctioned, causing promotional and social emails to land in Primary inboxes and some legitimate messages to be flagged with suspicious security warnings for users globally.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
Get the infrastructure, lures, and IOCs behind this campaign, ready to push into your email and identity stack.
4 references tracked. Mallory keeps watching after this page renders.
scworld.com
Open sourcescworld.com
Open sourcescworld.com
Open sourcetechrepublic.com
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Multiple campaigns are abusing *legitimate* cloud and platform workflows to make phishing and fraud harder to detect. Attackers are generating real Apple and PayPal invoice/dispute emails and embedding scam phone numbers in user-controlled fields (e.g., “seller notes”), resulting in messages that carry valid **DKIM** signatures and originate from high-reputation domains; this “**DKIM replay**” style abuse bypasses many email controls because authentication validates the sender domain, not the safety of the embedded content. In parallel, threat actors are leveraging free **Google Firebase** developer accounts to host brand-mimicking phishing pages on trusted `firebaseapp.com` / `web.app` subdomains, increasing delivery and click-through rates by exploiting domain reputation and common allowlisting of Google infrastructure. A separate but related social-engineering technique targets **Telegram** users by manipulating Telegram’s official authentication workflows to obtain fully authorized sessions rather than simply stealing passwords. Victims are lured to Telegram-lookalike pages (often on ephemeral domains) that prompt QR scanning or phone-number entry; user interaction triggers a real login attempt initiated by the attacker, and once the victim approves the authorization prompt on their device, the attacker gains persistent account access and can pivot to follow-on attacks via the victim’s contacts. These incidents collectively highlight a shift toward “living off trusted services,” where adversaries avoid compromising vendors and instead weaponize legitimate features, trusted domains, and sanctioned authentication flows to reduce detection and increase victim compliance.
Jun 29, 2026
Threat actors are increasingly improving phishing success rates by abusing *trusted* channels and infrastructure rather than relying on generic lures. One observed intrusion hijacked an active executive email thread via a compromised contractor account, allowing the attacker to reply inline with a link to a Microsoft 365 lookalike login flow; analysis of detonated samples indicated use of the **EvilProxy** adversary-in-the-middle phishkit, with layered anti-bot gating (e.g., Cloudflare Turnstile) and dynamic HTML/PDF content to capture credentials without exploiting software vulnerabilities. Separately, Rapid7 documented a cloud-abuse incident where attackers used **compromised AWS credentials** to stand up phishing/spam operations using **AWS WorkMail**, leveraging Amazon’s sender reputation and sidestepping typical **SES** anti-abuse controls while generating limited, service-native telemetry that can blend into normal administrative activity. A parallel, large-scale consumer fraud operation aligned with the **“PayTool”** ecosystem was reported targeting Canadian residents through SMS-driven lures (e.g., unpaid fines) that route victims through high-fidelity impersonations of the **Government of Canada**, **Air Canada**, and **Canada Post**, including province-selection workflows designed to mimic legitimate federal-to-provincial service handoffs before directing victims to localized scam domains. In contrast, LevelBlue SpiderLabs’ write-up is broader sector telemetry on education-targeted attacks (e.g., brute force `T1110`, credential dumping `T1003`, Kerberos ticket forgery `T1558`) and does not describe the same specific phishing/fraud campaigns, though it reinforces that credential theft remains a dominant initial access path across industries.
Jun 29, 2026
Security researchers reported multiple **phishing evasion** techniques designed to defeat modern email and AI-assisted defenses rather than relying only on traditional lure quality. One campaign analyzed by KnowBe4 used **graymail-style content padding** and extreme whitespace insertion to manipulate NLP-based email security tools, placing benign promotional text, legitimate signatures, and trusted links far below the visible phishing lure so scanners would weigh the message as less malicious. A separate LevelBlue-tracked trend showed attackers abusing enterprise **URL rewriting** and *Safe Links*-style protections by sending phishing through compromised accounts, causing security gateways to generate trusted wrapped URLs that could then be reused in campaigns targeting **Microsoft 365** users. The activity reflects a broader shift toward exploiting the gap between what users see and what automated systems inspect. In the URL-rewriting abuse, operators tied to **Tycoon2FA** and **Sneaky2FA** built multi-layer redirect chains across several trusted vendor domains to obscure final destinations and steal credentials and MFA session cookies through adversary-in-the-middle infrastructure, enabling account takeover, internal phishing, data theft, and sometimes ransomware follow-on activity. Related research from LayerX showed a different but thematically aligned evasion method in which **font rendering and CSS** make webpages display malicious commands to users while AI assistants parsing the underlying HTML see only harmless text, underscoring that attackers are increasingly targeting AI and trust-based inspection layers as part of phishing and social-engineering operations.
Jun 29, 2026