Microsoft Reportedly Blocks 1.6 Million Bot Signup Attempts Per Hour Amid Global AI-Driven Fraud Surge

Between April 2024 and April 2025, Microsoft reportedly blocked 1.6 million bot signups per hour and disrupted $4 billion in fraud attempts linked to AI-enhanced scams. The company's Cyber Signals report details how generative AI is being used to fabricate realistic e-commerce sites, job offers, customer service bots, and phishing lures. Fraud actors now automate mass-deceptive campaigns with fake reviews, deepfakes, and cloned brand domains at unprecedented scale and speed.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involves cybercriminals using generative AI, deepfakes, and voice cloning to create highly convincing fake storefronts and job listings to defraud individuals for financial gain.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The risk is caused by human cybercriminals intentionally deploying generative AI tools post-deployment to execute sophisticated fraud campaigns.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Risk Level 3: Limited Risk. The report describes the use of chatbots and AI-generated content such as deepfakes and voice cloning to run scams, which fall under the category of AI systems requiring transparency obligations.

AI system and alleged parties

• AI system: unspecified
• AI purpose: Chatbot; Voice Generation
• Behaviour type: Assistant
• Alleged developer: Various voice cloning technology developers, Various generative AI developers, Various deepfake technology developers
• Alleged deployer: Unknown scammers, Cyber criminal networks
• Alleged harmed parties: Microsoft, General public, Financial institutions, Enterprises, Consumers

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Substantial, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Minor, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Minor, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Financial

Reported: The report explicitly describes massive financial losses, noting Microsoft thwarted $4 billion in fraud attempts globally, and Australians lost $9.8 million to shopping scams in 2024.

Directly caused: The report does not specify the exact direct losses from AI-scams alone, but notes Australians lost $9.8 million to shopping scams in 2024, averaging about $978 per reported occurrence.

Inferred additional harm: Given the $4 billion in thwarted fraud, actual successful global financial losses from these AI-powered scams likely run into hundreds of millions or billions of dollars.

Malicious content

Reported: The report explicitly describes the creation of malicious content, including fake e-commerce websites, deepfake influencer videos, and automated phishing emails.

Directly caused: Scammers directly created fake websites, AI-generated product reviews, cloned voices, and deepfake interviews to deceive victims.

Inferred additional harm: Millions of malicious emails, fake reviews, and deceptive ads were likely spread globally across social media and e-commerce platforms.

Privacy

Reported: The report explicitly describes privacy risks, noting that scammers use AI-generated interviews and automated phishing emails to trick candidates into sharing personal data.

Directly caused: Scammers targeted job seekers to trick them into sharing personal data, resumes, and bank account details.

Inferred additional harm: Thousands of victims likely had their personally identifiable information stolen and potentially sold on the dark web or used for identity theft.

Epistemic

Reported: The report explicitly describes epistemic harm through the creation of highly convincing fake websites, AI-generated product reviews, and fake job listings that erode trust in online ecosystems.

Directly caused: Scammers generated fake product descriptions, reviews, and testimonials to dupe consumers into believing they were interacting with genuine merchants.

Inferred additional harm: The widespread deployment of AI-generated fake content likely degrades the overall trustworthiness of the digital commerce and employment information ecosystems.

People affected

• Occurrences reported: 1
• People reportedly harmed: 10022
• People reportedly exposed: 10022

Potential causes

Management

• Reactive security approach: Security is often treated as an afterthought rather than a design requirement.
• Slow adoption of defense AI: Attackers adopt new technologies faster than defensive organizations can.

Technology

• Lowered technical bar for fraud: Generative AI makes creating believable scam content cheap and fast.
• Easily bypassed ethical algorithms: Scammers bypass safety filters by slightly rephrasing their prompts.
• AI-powered deception tools: Chatbots and deepfakes mimic real humans to stall chargebacks and fool victims.

Data Inputs

• Social media data scraping: Scammers scrape personal data to create highly targeted phishing lures.
• Stolen credentials on dark web: Stolen customer information is sold on the dark web to feed scam campaigns.
• Web scraping of company data: Scammers scrape company sites to build detailed profiles of employees.

Human Factors

• Convenience over safety priority: Users treat security as an afterthought to maintain online convenience.
• Insufficient digital literacy: Users lack the training to identify sophisticated AI scams and deepfakes.
• Vulnerability to psychological triggers: Scammers exploit trust in social proof, urgency, and fake reviews.

Process and Methods

• Scam-as-a-service availability: Unsophisticated scammers can purchase advanced kits on the dark web.
• Inadequate employer verification: Job platforms lack multi-factor authentication for employer accounts.
• Lack of proactive fraud controls: Historically, security controls were not embedded directly into product design.

Regulatory Environment

• Global jurisdiction challenges: Cybercriminals operate internationally, making cross-border enforcement hard.
• Lagging AI ethics guidelines: Ethical principles exist but lack strong legal enforcement mechanisms.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent, detailed statistics from Microsoft's Cyber Signals report regarding the scale of AI-driven fraud, including specific figures on thwarted fraud ($4 billion) and blocked bot signups (1.6 million/hour). The role of generative AI in enabling these scams is clearly explained, and there are no major conflicting accounts among the sources.
• Ambiguities identified: The exact proportion of the $4 billion in thwarted fraud that specifically involved generative AI versus traditional cybercrime methods is not fully disaggregated.
• Alternative interpretations: Some of the described activities, like Storm-1811's Quick Assist abuse, did not use AI, which could lead to overestimating the role of AI in all reported tech support scams.

The incident details a massive global surge in AI-driven cyber fraud and scams, leveraging deepfakes, voice cloning, and generative AI to target consumers and job seekers. While the financial scale is significant with 4 billion dollars in thwarted fraud, the national security implications are minor and primarily affect economic security and societal trust in digital platforms.

• Overall national security impact: Minor
• Scope: Multiple nations
• Primary target: No clear primary
• Other affected: Australia, United States, Germany, China
• Alleged perpetrator: Cybercriminals

Threat characteristics

• Imminence: Long-term. Represents an ongoing strategic capability and persistent threat vector rather than a sudden, time-critical national security crisis.
• Autonomy: Human-controlled. AI systems are used as productivity tools to generate content, but human cybercriminals control and orchestrate the overall campaigns.
• Novelty: Evolved capability. Represents a significant advancement of existing fraud and social engineering tactics through the integration of generative AI and deepfakes.

Impact by dimension

• Physical security: Negligible. No threat to physical systems, critical infrastructure, or human safety was reported in this incident.
• Information security: Minor. Involves widespread use of deepfakes and voice cloning, but these are leveraged for financial fraud rather than state-level intelligence compromise or political information warfare.
• Sovereignty: Negligible. No disruption to state authority, electoral systems, or core government decision-making processes was reported.
• Economic security: Minor. Massive scale of AI-enabled financial fraud, with 4 billion dollars in thwarted attempts, representing a minor but persistent threat to economic security and consumer trust.
• Societal stability: Minor. Widespread deployment of AI-generated fake content and fake job listings targeting individuals, causing minor epistemic harm and eroding trust in digital commerce.