Microsoft is sounding the alarm on the rising threat of AI-driven scams, revealing in its latest Cyber Signals report that it blocked $4 billion in fraud attempts over the past year. The company also intercepted approximately 1.6 million bot sign-ups every hour, underlining the scale and speed of the escalating threat.

The ninth edition of the report, AI-powered deception: Emerging Fraud Threats and Countermeasures, warns that generative AI has drastically lowered the technical barrier for cybercriminals, allowing them to launch convincing scams in minutes instead of days.

“Cybercrime is a trillion-dollar problem, and it’s been going up every year for the past 30 years,” says Kelly Bissell, Corporate Vice President of Anti-Fraud and Product Abuse at Microsoft. “Now we have AI that can make a difference at scale and help us build security and fraud protections into our products much faster.”

Fraudsters are increasingly using AI to scrape corporate data, create fake online stores and reviews, and run employment scams with AI-generated job listings and phishing campaigns. The report notes a surge in fraudulent activity originating from China and Germany, the latter due to its prominence in e-commerce.

To counter the rise of AI-powered fraud, Microsoft has strengthened its security arsenal. Microsoft Defender for Cloud monitors Azure environments, Microsoft Edge uses deep learning to block phishing sites and Windows Quick Assist now alerts users to potential tech-support scams — blocking over 4,400 suspicious attempts daily.

Under its Secure Future Initiative, Microsoft now mandates that all product teams integrate fraud prevention measures from the ground up, ensuring systems are “fraud-resistant by design.”

Microsoft urges users to remain vigilant—verifying sources, avoiding urgency traps, and safeguarding sensitive data—while encouraging enterprises to implement multi-factor authentication and deploy deepfake detection tools to stay ahead of evolving threats.