Krux
May 1, 2026
ChatGPT Users Send 15M Scam Checks Monthly
Published: May 1, 2026 at 12:15 AM
Updated: May 1, 2026 at 12:15 AM
100-word summary
OpenAI just published a five-part plan to hand AI defense tools to government agencies and businesses while keeping the most dangerous capabilities locked down. The centerpiece is Trusted Access for Cyber, a tiered program that will give verified defenders advanced AI tools, starting with federal agencies and the financial sector. Smaller outfits get access through middlemen like security providers. Here's the surprise: regular ChatGPT users already send over 15 million messages per month asking it to check potential scams. That's more consumer-level cyber paranoia than most security companies see in a year, suggesting everyday defenses may matter as much as enterprise fortifications.
What happened
OpenAI just published a five-part plan to hand AI defense tools to government agencies and businesses while keeping the most dangerous capabilities locked down. The centerpiece is Trusted Access for Cyber, a tiered program that will give verified defenders advanced AI tools, starting with federal agencies and the financial sector. Smaller outfits get access through middlemen like security providers. Here's the surprise: regular ChatGPT users already send over 15 million messages per month asking it to check potential scams.
Why it matters
That's more consumer-level cyber paranoia than most security companies see in a year, suggesting everyday defenses may matter as much as enterprise fortifications.