A broad coalition of more than 100 leading global companies, including OpenAI, Anthropic, Microsoft, Google, and Amazon Web Services, has issued an urgent warning: the wave of cyberattacks using artificial intelligence in the coming months will become not just more frequent, but qualitatively different—significantly more sophisticated and destructive.
This is an unprecedented step, with competitors uniting in the face of a common threat. Signatories to the collective statement include representatives not only from the technology sector, but also from banking, insurance companies, and semiconductor manufacturers. Among the signatories are CrowdStrike, Cloudflare, and Palo Alto Networks, underscoring the systemic nature of the problem.
Four Fronts of Defense
In my analysis, I identify four key areas where the authors demand immediate action. First, cybersecurity must be elevated to the level of top company management—this is no longer an IT department issue, but a matter of business survival. Second, the most critical vulnerabilities that have accumulated in corporate systems for decades must be urgently addressed.
Special attention is given to critical infrastructure operators. Solution providers must make AI protection accessible to hospitals, utilities, and local government bodies, which often remain defenseless against modern threats. Finally, leading AI labs must open access to their models and fund the training of defense specialists.
"Modern advances in AI are already giving defenders new ways to eliminate weaknesses that have accumulated over the years. If we act decisively, we can use this window of opportunity to make the digital world much safer," the statement emphasizes.
Alarming Numbers
My own research confirms: these are not theoretical concerns. Taiwan's TeamT5 has recorded that Chinese hacker groups have doubled the number of attacks after switching to DeepSeek and other open models. The reason is not power, but accessibility—attackers choose cheap tools.
Anthropic's data for June is particularly telling: an analysis of 832 blocked accounts revealed that the share of attackers with medium and high risk grew from 33% to 56% in just one year. Moreover, AI now independently handles privilege escalation and lateral network movement—tasks that previously required deep technical skills. The level of attack has ceased to depend on the attacker's experience.
The North Korean group Kimsuky is actively testing local AI tools, while TRM Labs analysts rate the use of AI for criminal purposes at 54 out of 100—nearly double the 28 points in 2024. In the first half of 2026, 201 hacks of crypto projects were recorded, compared to 83 a year earlier.
My verdict: we stand on the threshold of a new era of digital warfare, where AI becomes a double-edged sword. The current leap in technological development is our last chance to build reliable defenses before the attackers' arsenal becomes truly lethal. The crypto industry, as the most vulnerable target, should heed this signal especially carefully.