AI-Driven Cyberattacks Quadruple Since 2024, Threatening 100 Million People
Updated
Updated · The Atlantic · Jun 16
AI-Driven Cyberattacks Quadruple Since 2024, Threatening 100 Million People
3 articles · Updated · The Atlantic · Jun 16
Summary
Palo Alto Networks recorded a fourfold jump in daily attacks from 2024 to 2025 as AI tools let hackers write malware faster, automate espionage and exploit stolen data in minutes.
44 days is now the average time to weaponize a known software flaw, down from more than 700 days in 2020 and faster than many security teams can patch.
Critical systems are increasingly exposed: Canvas outages hit thousands of schools last month, ShinyHunters allegedly breached an Oracle HR system affecting 100-plus organizations, and Meta said a failure exposed about 30,000 Instagram accounts.
Mozilla used Anthropic's Mythos to fix 400-plus Firefox bugs in April—about 20 times its usual monthly pace—but experts say defensive adoption is late and public access to top cyber models has been curtailed.
2026 could bring blackouts, telecom breaches and banking losses, experts warn, with hospitals, utilities and smaller banks especially vulnerable because legacy software and thin IT budgets leave little margin for missed flaws.
As AI supercharges cyberattacks, are global security responses falling dangerously behind the escalating threat?
AI can now autonomously cause catastrophic damage. Who will be held accountable when it inevitably goes wrong?
The Global AI Cybercrime Crisis: 600 Million Daily Attacks and the $40 Billion Threat
Overview
The report highlights the alarming rise of AI-driven cyberattacks, with the digital world facing around 600 million attacks daily and nearly 54 victims every second. Nearly 60% of businesses have suffered ransomware attacks this year, while AI-enabled fraud is rapidly increasing, especially through deepfake audio, images, and videos. In the Asia-Pacific region, AI-related fraud attempts surged by 194% in 2024, with many incidents involving malicious voice and video calls. The growing use of deepfake vishing and the involvement of nation-state actors underscore the urgent need for stronger defenses and global cooperation against these evolving threats.