Phish-Proof My Inbox
A consumer app that screens incoming emails and SMS for the specific attack patterns used in recent AI-driven phishing campaigns, warning the user before they click.
Older adults and busy parents who handle their own email
- Inline explainer showing why a message was flagged ('the sender's domain was registered 3 days ago')
- Family-shield plan so one adult can monitor a parent's inbox
- Weekly recap of blocked attempts so users see the value
- One-tap report that forwards suspect messages to the FTC/Canadian Anti-Fraud Centre
AI phishing is now sophisticated enough to fool consumers (per the CNBC article), and high-profile incidents create a teachable moment when attention is already captured.
AI-driven phishing has exploded: phishing surged 1,265% due to generative AI and 82.6% of phishing emails now use AI language models, with 53.5% YoY growth — a clear and rapidly expanding threat base for the target users.AI Cybersecurity Statistics in 2025: Comprehensive Data on Threats ↗AI Cyber Threat Statistics: The 2025 Landscape of AI-Powered Cyberattacks ↗
Enterprise email-security space is crowded (Barracuda, KnowBe4, Proofpoint, and 10+ listed anti-phishing tools) and Gmail/Apple/Microsoft are adding built-in phishing detection, leaving limited differentiated room for a consumer-only AI-pattern screener.10 Best Anti-Phishing Tools in 2026 ↗Top 12 Email Filtering Solutions for Enhanced Inbox Security ↗
Consumers do pay for security suites (Norton/McAfee ~$50–100/yr), but Gmail/iOS already ship free phishing warnings and a standalone consumer-only AI-phishing app has narrower willingness-to-pay; monetization is workable but not premium.Top 10 Phishing Protection Alternatives 2026 ↗AI Cybersecurity Statistics in 2025 ↗
AI phishing is structurally accelerating — 1,265% surge, 53.5% YoY increase, and the underlying generative-AI capability curve continues rising, so the threat (and need for detectors) is durable for years.AI Cyber Threat Statistics: The 2025 Landscape ↗Phishing Threat Trends Report 2025 (KnowBe4) ↗
Feasible to build using LLM-based classifiers on Gmail API/IMAP + SMS forwarding, but requires ongoing model retraining as attack patterns evolve and must coexist with platform-native protections; moderate complexity, not trivial.Top 12 Email Filtering Solutions for Enhanced Inbox Security ↗