

The Week in Email Marketing
Inbox Intel: from emailexpert
emailexpert | Rating 0 (0) (0) |
https://emailexpert.com | Launched: Jul 20, 2025 |
podcasts@emailexpert.org | Season: 1 Episode: 10 |
Here are the show notes for this podcast episode, "Emailexpert Inbox Intel," covering critical topics in email marketing and security:
Email Expert Insights: Navigating AI Threats, Legal Minefields, and Data Pitfalls
This episode of Email Expert Insights dives deep into the most critical challenges and groundbreaking developments shaping the email landscape today, from sophisticated AI-driven cyber threats to costly legal battles and the often-overlooked problem of dirty data.
I. New Class of AI-Driven Email Threats: The Google Gemini Exploit
• AI-Driven Phishing: Cybercriminals are now leveraging AI tools not just to generate attacks, but to weaponize the recipient's own inbox AI features. A new vulnerability in Gmail's Gemini summarization feature highlights this emerging threat.
• Prompt Injection Explained: This novel phishing technique bypasses traditional email defenses by exploiting Google Workspace's AI-powered summarizer, Gemini. It requires no links, no attachments, and no visible malicious content in the email body. Instead, it relies on "prompt injection," hidden invisibly within the email, often using techniques like hidden HTML/CSS styling, to manipulate the Gemini summary.
• Demonstrated Exploit: In a real-world example, the Gemini summary falsely warned, "Gemini has detected your Gmail password has been compromised, please call us immediately at [phone number]". This "carefully crafted hallucination" is a social engineering tactic designed to induce panic and an immediate response, allowing the malicious AI summary to deliver the attack.
• Shift in Attack Surface: This incident signals a significant shift where attackers are designing exploits that target how machines interpret email content, rather than how humans do. AI-generated UI elements like summaries, alerts, and previews are now attack surfaces in their own right.
• Mitigation and Future Outlook: Google has acknowledged the issue and is "hardening its protections against prompt injection attacks". For security teams, this calls for improved input sanitization in AI summarization engines, new heuristics to detect prompt injection attempts, and enhanced user education about relying on AI-generated summaries for security-related information. Legitimate senders should also maintain clarity and consistency in message formatting and monitor inbox renderings.
II. Costly Legal Pitfalls in Email Marketing
• Nike Inc. Class Action Lawsuit: Nike is facing a proposed class action lawsuit in Washington state for using misleading subject lines that allegedly created a false sense of urgency. Examples include "Only a few hours left" or "Ends tonight," which implied imminent sale endings but promotions were reportedly extended or fabricated. The lawsuit claims violations of the Washington Commercial Electronic Mail Act (CEMA) and the Washington Consumer Protection Act.
• Crucial Legal Precedent - Brown v. Old Navy, LLC: This Nike case is significantly bolstered by a landmark Washington Supreme Court ruling in April 2025 in Brown v. Old Navy, LLC. This ruling broadly interpreted CEMA to impose a $500 statutory penalty on every commercial email containing false or misleading information in its subject line sent to Washington residents, without requiring proof of actual financial damages. The "injury is receiving the email that violates CEMA".
• Tim Hortons Class Action Lawsuit: A Quebec Superior Court judge authorized a class action lawsuit against Tim Hortons due to a "catastrophic email marketing error" in April 2024. Approximately 500,000 contest participants, including thousands in Quebec, falsely received emails stating they had won a $64,000 boat and trailer. Follow-up emails retracted the win, citing "technical issues".
• Quebec's Consumer Protection Act: The lawsuit argues that Quebec's Consumer Protection Act prevents companies from simply claiming "mistake" to void contractual agreements formed by contest win notifications.
• Lessons for Marketers: These cases highlight the importance of accuracy in subject lines (avoiding fabricated scarcity), geo-targeting compliance with specific state laws, utilizing dynamic content for timely updates, and conducting regular audits and legal reviews. The Tim Hortons incident further emphasizes the need for robust testing protocols, approval workflows, and pre-planned crisis communication templates for contest and promotional emails.
III. The Silent Saboteur: Dirty Data
• Pervasive Problem: A new report, "The State of CRM Data Management in 2025," reveals that 76% of companies admit less than half of their CRM data is accurate or complete. Furthermore, 37% attribute lost revenue directly to poor data quality through mistargeted campaigns, missed follow-ups, and distorted reporting.
• Financial Impact: Bad data is estimated to consume 15% of annual revenue (Gartner). An IBM figure places the U.S. cost of poor data quality at a staggering $3.1 trillion annually. Data scientists spend approximately 60% of their time cleaning data rather than extracting insights.
• AI Amplifies Issues: When AI tools are trained on faulty inputs, they lead to "broken logic chains," "misfiring triggers," and "personas built on outdated job titles". As AI amplifies everything, it also magnifies the problems caused by bad data.
• Pragmatic Steps for Email Professionals: To combat dirty data, professionals should run data audits to identify issues, enforce input standards using validation rules and tools like Bouncer or ZeroBounce, and automate data hygiene with tools like Dedupely and Openprise. Strategic data enrichment, prioritizing deliverability with clean lists, and aligning AI plans with verified data integrity are also crucial.
IV. Email Security Under Scrutiny: Ireland's NTMA Phishing Loss
• Significant Loss: Ireland's National Treasury Management Agency (NTMA) reportedly suffered losses of up to €5 million due to a "sophisticated, multi-layered phishing attack" that targeted staff with fraudulent payment requests.
• DMARC Configuration Weakness: While the ntma.ie
domain had a DMARC record, it was configured with a p=none policy. This means it would check for failures and report them, but still allow spoofed emails to pass through and potentially land in staff inboxes, offering "little real-world protection".
• Foundational Components Missing: Additionally, SPF and DKIM, which are the foundational components DMARC relies on for authentication, appeared to be misconfigured or absent for ntma.ie
.
• Importance of Proper Implementation: This incident reinforces the critical importance of properly implemented DMARC, SPF, and DKIM to significantly reduce the risk of domain spoofing. A phased approach for DMARC implementation is recommended: starting with p=none
for monitoring, then gradually transitioning to p=quarantine
, and finally p=reject
only when confident all legitimate email is authenticated.
• Public Sector Readiness: The incident also highlights concerns about email readiness in the public sector, which often lags commercial entities in adopting best-practice authentication policies, despite new requirements from bulk senders like Gmail, Yahoo, and Microsoft. The NTMA's permissive email authentication posture was "avoidable" and underscores that a foundational DMARC policy is just the first step in a comprehensive email security strategy.
V. Industry Excellence: Jay Oram Honored
• David Baker Lifetime Achievement Award: Jay Oram, Head of Development at ActionRocket, was presented with the 2025 David Baker Lifetime Achievement Award at the ANA Email Excellence Center (EEC) awards ceremony on July 17, 2025.
• Recognizing Contributions: This award, renamed in 2024 to commemorate pioneering CRM executive David Baker, recognizes vendor-side practitioners whose careers have advanced the craft and community of email marketing.
• Innovator in Email: Oram was specifically honored for more than a decade of pushing the boundaries of interactive, AMP, and live-data email. His work has powered campaigns for major brands, and his code tutorials are widely used.
• Future Plans: Oram has teased an upcoming webinar conversation with strategist Ryan Phelan to discuss lessons and the future of kinetic email, and hinted at new accessibility tooling slated for open-source release this autumn.
Think of navigating the world of email marketing and security as sailing a ship through increasingly stormy seas. AI-driven threats are like new, unseen icebergs that AI-powered radar systems might mistake for safe passage. Legal pitfalls are the shifting sands and hidden reefs that can ground your vessel without warning if you don't understand the local charts. Dirty data is the barnacles on your hull, silently slowing you down and wasting fuel, even if your engines (AI strategies) are powerful. And proper email authentication is your ship's sturdy hull and reliable navigation systems, ensuring your communications reach their destination safely and aren't impersonated by pirates. Every part must be robust and well-maintained to avoid catastrophe.
#EmailMarketing #B2BMarketing #DataPrivacy #MarketingTechnology #EmailDeliverability #AIMarketing
SUBSCRIBE
Episode Chapters

Here are the show notes for this podcast episode, "Emailexpert Inbox Intel," covering critical topics in email marketing and security:
Email Expert Insights: Navigating AI Threats, Legal Minefields, and Data Pitfalls
This episode of Email Expert Insights dives deep into the most critical challenges and groundbreaking developments shaping the email landscape today, from sophisticated AI-driven cyber threats to costly legal battles and the often-overlooked problem of dirty data.
I. New Class of AI-Driven Email Threats: The Google Gemini Exploit
• AI-Driven Phishing: Cybercriminals are now leveraging AI tools not just to generate attacks, but to weaponize the recipient's own inbox AI features. A new vulnerability in Gmail's Gemini summarization feature highlights this emerging threat.
• Prompt Injection Explained: This novel phishing technique bypasses traditional email defenses by exploiting Google Workspace's AI-powered summarizer, Gemini. It requires no links, no attachments, and no visible malicious content in the email body. Instead, it relies on "prompt injection," hidden invisibly within the email, often using techniques like hidden HTML/CSS styling, to manipulate the Gemini summary.
• Demonstrated Exploit: In a real-world example, the Gemini summary falsely warned, "Gemini has detected your Gmail password has been compromised, please call us immediately at [phone number]". This "carefully crafted hallucination" is a social engineering tactic designed to induce panic and an immediate response, allowing the malicious AI summary to deliver the attack.
• Shift in Attack Surface: This incident signals a significant shift where attackers are designing exploits that target how machines interpret email content, rather than how humans do. AI-generated UI elements like summaries, alerts, and previews are now attack surfaces in their own right.
• Mitigation and Future Outlook: Google has acknowledged the issue and is "hardening its protections against prompt injection attacks". For security teams, this calls for improved input sanitization in AI summarization engines, new heuristics to detect prompt injection attempts, and enhanced user education about relying on AI-generated summaries for security-related information. Legitimate senders should also maintain clarity and consistency in message formatting and monitor inbox renderings.
II. Costly Legal Pitfalls in Email Marketing
• Nike Inc. Class Action Lawsuit: Nike is facing a proposed class action lawsuit in Washington state for using misleading subject lines that allegedly created a false sense of urgency. Examples include "Only a few hours left" or "Ends tonight," which implied imminent sale endings but promotions were reportedly extended or fabricated. The lawsuit claims violations of the Washington Commercial Electronic Mail Act (CEMA) and the Washington Consumer Protection Act.
• Crucial Legal Precedent - Brown v. Old Navy, LLC: This Nike case is significantly bolstered by a landmark Washington Supreme Court ruling in April 2025 in Brown v. Old Navy, LLC. This ruling broadly interpreted CEMA to impose a $500 statutory penalty on every commercial email containing false or misleading information in its subject line sent to Washington residents, without requiring proof of actual financial damages. The "injury is receiving the email that violates CEMA".
• Tim Hortons Class Action Lawsuit: A Quebec Superior Court judge authorized a class action lawsuit against Tim Hortons due to a "catastrophic email marketing error" in April 2024. Approximately 500,000 contest participants, including thousands in Quebec, falsely received emails stating they had won a $64,000 boat and trailer. Follow-up emails retracted the win, citing "technical issues".
• Quebec's Consumer Protection Act: The lawsuit argues that Quebec's Consumer Protection Act prevents companies from simply claiming "mistake" to void contractual agreements formed by contest win notifications.
• Lessons for Marketers: These cases highlight the importance of accuracy in subject lines (avoiding fabricated scarcity), geo-targeting compliance with specific state laws, utilizing dynamic content for timely updates, and conducting regular audits and legal reviews. The Tim Hortons incident further emphasizes the need for robust testing protocols, approval workflows, and pre-planned crisis communication templates for contest and promotional emails.
III. The Silent Saboteur: Dirty Data
• Pervasive Problem: A new report, "The State of CRM Data Management in 2025," reveals that 76% of companies admit less than half of their CRM data is accurate or complete. Furthermore, 37% attribute lost revenue directly to poor data quality through mistargeted campaigns, missed follow-ups, and distorted reporting.
• Financial Impact: Bad data is estimated to consume 15% of annual revenue (Gartner). An IBM figure places the U.S. cost of poor data quality at a staggering $3.1 trillion annually. Data scientists spend approximately 60% of their time cleaning data rather than extracting insights.
• AI Amplifies Issues: When AI tools are trained on faulty inputs, they lead to "broken logic chains," "misfiring triggers," and "personas built on outdated job titles". As AI amplifies everything, it also magnifies the problems caused by bad data.
• Pragmatic Steps for Email Professionals: To combat dirty data, professionals should run data audits to identify issues, enforce input standards using validation rules and tools like Bouncer or ZeroBounce, and automate data hygiene with tools like Dedupely and Openprise. Strategic data enrichment, prioritizing deliverability with clean lists, and aligning AI plans with verified data integrity are also crucial.
IV. Email Security Under Scrutiny: Ireland's NTMA Phishing Loss
• Significant Loss: Ireland's National Treasury Management Agency (NTMA) reportedly suffered losses of up to €5 million due to a "sophisticated, multi-layered phishing attack" that targeted staff with fraudulent payment requests.
• DMARC Configuration Weakness: While the ntma.ie
domain had a DMARC record, it was configured with a p=none policy. This means it would check for failures and report them, but still allow spoofed emails to pass through and potentially land in staff inboxes, offering "little real-world protection".
• Foundational Components Missing: Additionally, SPF and DKIM, which are the foundational components DMARC relies on for authentication, appeared to be misconfigured or absent for ntma.ie
.
• Importance of Proper Implementation: This incident reinforces the critical importance of properly implemented DMARC, SPF, and DKIM to significantly reduce the risk of domain spoofing. A phased approach for DMARC implementation is recommended: starting with p=none
for monitoring, then gradually transitioning to p=quarantine
, and finally p=reject
only when confident all legitimate email is authenticated.
• Public Sector Readiness: The incident also highlights concerns about email readiness in the public sector, which often lags commercial entities in adopting best-practice authentication policies, despite new requirements from bulk senders like Gmail, Yahoo, and Microsoft. The NTMA's permissive email authentication posture was "avoidable" and underscores that a foundational DMARC policy is just the first step in a comprehensive email security strategy.
V. Industry Excellence: Jay Oram Honored
• David Baker Lifetime Achievement Award: Jay Oram, Head of Development at ActionRocket, was presented with the 2025 David Baker Lifetime Achievement Award at the ANA Email Excellence Center (EEC) awards ceremony on July 17, 2025.
• Recognizing Contributions: This award, renamed in 2024 to commemorate pioneering CRM executive David Baker, recognizes vendor-side practitioners whose careers have advanced the craft and community of email marketing.
• Innovator in Email: Oram was specifically honored for more than a decade of pushing the boundaries of interactive, AMP, and live-data email. His work has powered campaigns for major brands, and his code tutorials are widely used.
• Future Plans: Oram has teased an upcoming webinar conversation with strategist Ryan Phelan to discuss lessons and the future of kinetic email, and hinted at new accessibility tooling slated for open-source release this autumn.
Think of navigating the world of email marketing and security as sailing a ship through increasingly stormy seas. AI-driven threats are like new, unseen icebergs that AI-powered radar systems might mistake for safe passage. Legal pitfalls are the shifting sands and hidden reefs that can ground your vessel without warning if you don't understand the local charts. Dirty data is the barnacles on your hull, silently slowing you down and wasting fuel, even if your engines (AI strategies) are powerful. And proper email authentication is your ship's sturdy hull and reliable navigation systems, ensuring your communications reach their destination safely and aren't impersonated by pirates. Every part must be robust and well-maintained to avoid catastrophe.
#EmailMarketing #B2BMarketing #DataPrivacy #MarketingTechnology #EmailDeliverability #AIMarketing
Welcome to Emailexpert Inbox Intel, your essential guide to navigating the complex and ever-evolving world of email marketing and security. In this pivotal podcast episode, we uncover the most pressing challenges and groundbreaking developments impacting the industry today.
We delve into a new class of AI-driven threats, exemplified by a novel phishing exploit against Google Workspace's Gemini summarization feature. Learn how cybercriminals are now weaponizing recipients' own inbox AI by embedding invisible "prompt injection" techniques in emails, causing AI summaries to generate false warnings like "Gemini has detected your Gmail password has been compromised". This sophisticated tactic bypasses traditional email defenses, highlighting a critical shift where attackers target how machines interpret email content, not just humans. We discuss the urgent need for improved input sanitization in AI summarization engines, new heuristics to detect prompt injection attempts, and enhanced user education about relying on AI-generated summaries, especially for security-related information.
The episode also shines a light on costly legal pitfalls in email marketing. Discover why Nike Inc. is facing a proposed class action lawsuit in Washington state for using misleading subject lines such as “Only a few hours left” or “Ends tonight,” which allegedly misrepresented sale durations. This case is significantly bolstered by a landmark Washington Supreme Court ruling in Brown v. Old Navy, LLC (April 2025), which broadly interpreted the Washington Commercial Electronic Mail Act (CEMA) to impose a $500 statutory penalty on every commercial email containing false or misleading information in its subject line sent to Washington residents, crucially without requiring proof of actual financial damages. Similarly, we examine the Tim Hortons class action lawsuit in Quebec, Canada, where a "catastrophic email marketing error" falsely informed 500,000 contest participants they had won a $64,000 boat and trailer. This incident underscores the severe legal and reputational risks of automation failures and the importance of understanding regional consumer protection laws, as Quebec's Consumer Protection Act prevents companies from simply claiming "mistake" to void contractual agreements formed by contest win notifications.
Furthermore, we confront the silent saboteur of brilliant AI marketing strategies: dirty data. A new report reveals that 76% of companies admit less than half of their CRM data is accurate or complete, directly costing them revenue through mistargeted campaigns, missed follow-ups, and distorted performance reporting. Gartner estimates bad data eats up 15% of annual revenue, and a figure from IBM places the U.S. cost of poor data quality at a staggering $3.1 trillion annually. Data scientists still spend around 60% of their time cleaning data instead of extracting insights from it. This problem is amplified when AI tools are trained on faulty inputs, leading to "broken logic chains" and "misfiring triggers". We provide pragmatic steps for email professionals, including running data audits, enforcing input standards, automating data hygiene with tools like Dedupely and Openprise, strategic data enrichment, and prioritizing deliverability, emphasizing that data quality is a "frontline revenue engine".
Finally, we analyze a €5 million phishing loss suffered by Ireland's National Treasury Management Agency (NTMA), which highlights critical weaknesses in email authentication. Despite having a DMARC record, the ntma.ie
domain was configured with p=none
, meaning it would check for failures and report them but still allow spoofed emails to pass through and potentially land in staff inboxes. This incident reinforces the operational importance of properly implemented DMARC, SPF, and DKIM (the foundational components DMARC relies on) to dramatically reduce the risk of domain spoofing, a tactic at the heart of many successful phishing campaigns. We discuss the recommended phased approach for DMARC implementation, moving from p=none
to p=quarantine
and eventually p=reject
. The incident also raises concerns about email readiness in the public sector, which often lags behind commercial entities in adopting best-practice authentication policies, despite new requirements from bulk senders like Gmail, Yahoo, and Microsoft.
We'll also take a moment to celebrate industry excellence, recognizing Jay Oram, Head of Development at ActionRocket, who was honored with the 2025 David Baker Lifetime Achievement Award on July 17, 2025, for his decade-plus contributions to advancing the craft and community of email marketing, particularly in interactive, AMP, and live-data email.
Join us to gain a deeper understanding of these crucial aspects of email, ensuring your strategies are robust, compliant, and genuinely effective.
Think of building a strong email presence in the digital age like constructing a secure, high-speed bridge: you need not only a robust design (your marketing strategy) and reliable materials (clean data), but also vigilant security checkpoints (authentication protocols) and a thorough understanding of the terrain's legal boundaries (compliance). Overlooking any single component, from a loose bolt to an unseen sinkhole, can lead to catastrophic failures, wasting resources and eroding trust.
#EmailMarketing #B2BMarketing #DataPrivacy #MarketingTechnology #EmailDeliverability #AIMarketing