As Criminals Innovate With AI, Cyber Defenses Scramble to Keep Up
The rapid advances in artificial intelligence are reshaping the battlefield between cybercriminals and security teams—raising the stakes for corporations and government agencies already struggling to keep threats at bay.
According to new research from cybersecurity firm Check Point Software Technologies, digital crime syndicates and ransomware gangs are exploiting generative AI tools to streamline and scale up attacks ranging from phishing campaigns to malware development. The result: not only are attacks growing more sophisticated, but the volume and velocity of incidents are increasing as bad actors leverage AI to automate key parts of their operations.
“The barrier to entry for cybercrime has never been lower,” said Sergey Shykevich, threat intelligence group manager at Check Point. “We’re seeing even novice attackers use AI tools to craft convincing phishing emails and malware that would have required programming expertise just a year ago.”
Check Point, whose research tracks global cyberthreats, says generative AI—such as large language models—has enabled an explosion in the creation of deepfake images, doctored audio and realistic-sounding messages, all designed to trick victims into clicking malicious links or surrendering sensitive credentials. By using AI, attackers can quickly customize lures to mirror a target organization’s unique vocabulary, internal references, or digital branding.
The company cited multiple cases, including AI-powered phishing attempts that mimicked C-suite executives’ voices to convince employees to authorize fraudulent wire transfers—tactics that have led to seven-figure losses in recent months. In one scenario, attackers harnessed public data and AI to impersonate a CEO on a video call, officials said.
Corporate Security at Crossroads
Meanwhile, security professionals face mounting pressure to stem the tide amid tight budgets and a persistent shortage of skilled cybersecurity workers—a gap some experts fear could widen if AI continues to favor criminals over defenders.
“The problem is not just that generative AI makes attacks more creative or convincing, but that it can do so at a scale and speed that overwhelms even well-resourced teams,” said Patrick Tiquet, vice president of security and architecture at Keeper Security, a Chicago-based password management firm.
The double-edged sword of AI, industry leaders add, is that defensive tools are also rapidly improving. Security vendors are integrating AI into detection engines, threat intelligence systems, and network monitoring tools to flag suspicious activity in real time and sift through mountains of code for vulnerabilities.
However, Tiquet cautions that implementation is not always straightforward. “AI is only as good as the data it’s trained on and the humans guiding it,” he said, saying that security teams must keep refining their models and guard against false positives or AI-generated blind spots. “The arms race is escalating.”
Data Privacy Risks Multiply
As business leaders clamor to adopt new generative AI solutions, experts warn that a parallel crisis is unfolding: the risk of accidental data leaks and privacy violations as employees feed proprietary information into chatbots, code s, or image generators that may not be adequately secured.
Check Point reported a surge in “shadow AI” use—applications deployed or accessed by employees without security teams’ oversight—leading to sensitive records or intellectual property being exposed to external servers or malicious actors. “Most organizations underestimate how easily confidential data can be leaked through unsanctioned AI platforms,” Shykevich said.
AI governance frameworks remain relatively immature, and many organizations are still crafting policies to determine what types of data users can share with cloud-based AI applications. Misconfigurations and lack of visibility, Check Point warns, could leave companies open to regulatory sanctions or intellectual property theft.
Regulators Eye New Rules
The growing concern over AI-driven cybercrime has caught the attention of regulators in Washington and Brussels, where policymakers are weighing new rules that would require stricter oversight for AI providers and their business clients. The Biden administration, along with the European Union, has signaled interest in mandating transparency, security assessments, and incident reporting for AI-enabled services.
Security executives say that while regulation can play a role, organizations should not wait for governments to act. “Companies need to immediately assess the AI exposure in their environments and implement clear usage policies for employees,” Tiquet said.
As one security leader put it: “It’s clear AI is here to stay on both sides of this arms race. The question now is who will adapt fastest.”