StrongestLayer Inc. has introduced two innovative products designed to enhance email security by verifying payment and payroll requests and preventing hidden AI manipulation commands from reaching corporate assistants like Microsoft 365 Copilot.
- Intent Workflows halts risky payment requests for verification to prevent fraud
- AI Instruction Intercept detects and blocks hidden AI manipulation commands
- Solutions available now for Microsoft 365 and Google Workspace environments
What happened
StrongestLayer Inc., an AI-focused email security startup, has launched two complementary tools designed to enhance protection against financial fraud and AI assistant manipulation via email. The first, Intent Workflows, places holds on payment, payroll, and executive account change requests to enable verification before execution. This feature aims to catch fraudulent requests that evade traditional security checks and appear legitimate, such as hijacked vendor emails requesting bank detail updates.
The second product, AI Instruction Intercept, protects AI assistants integrated with email, like Microsoft 365 Copilot and Google Gemini, from malicious prompt injections hidden in messages. It analyzes every email for concealed instructions in the context of the customer’s business and quarantines suspicious messages before reaching the assistant. Both offerings are compatible with Microsoft 365 and Google Workspace and designed to work alongside existing security solutions.
Why it matters
While only a small fraction of email attacks involve payment or bank detail changes, those represent a disproportionately high share of financial risk. StrongestLayer’s analysis showed that 3.6% of attacks were payment-related yet accounted for over 98% of estimated exposure when applying the Factor Analysis of Information Risk model. Traditional sender authentication methods fail against these attacks because a majority of fraudulent payment requests pass them unchecked.
Moreover, as AI assistants become embedded in business workflows, attackers are increasingly exploiting email to inject deceptive commands that could lead to sensitive data leaks or unwanted actions. By focusing on the intent behind financial requests rather than just message appearance, and by blocking covert AI instructions early, StrongestLayer’s new tools address critical gaps in email security mechanisms.
What to watch next
Adoption of Intent Workflows and AI Instruction Intercept by enterprises using Microsoft 365 and Google Workspace will provide an indicator of how urgently organizations prioritize prevention of financial fraud and AI prompt injection through email. Early feedback on usability—such as balancing verification holds against workflow efficiency—and effectiveness at catching complex fraud attempts will be key.
Additionally, the evolving threat landscape may spur further innovation in AI-driven email security, with competitors potentially launching similar products. Monitoring how attackers adapt to these new defenses, and whether prompt injection attempts evolve to bypass AI Instruction Intercept, will be important for assessing the long-term impact of StrongestLayer’s solutions.