The largest players in the technology industry, including OpenAI, Anthropic, Microsoft, Google, and Amazon Web Services, have joined forces in an unprecedented initiative. More than 100 companies have signed a collective statement warning of an imminent wave of cyberattacks using artificial intelligence, which will become not only larger in scale but also qualitatively more sophisticated.
What the industry is warning about
This document is not just a declaration of intent, but a clear action plan for four key groups. Corporations are strongly urged to elevate cybersecurity issues to the level of senior management and first and foremost close the most critical vulnerabilities. Security solution providers are urged to make AI protection accessible to critical infrastructure operators. Government agencies are urged to coordinate response measures and provide support to hospitals, utilities, and local administrations.
Special emphasis is placed on leading AI companies: they are invited to open access to their models and fund training for those who lack the resources for defense. The authors emphasize: "Modern advances in AI already give defenders new ways to eliminate weaknesses that have accumulated over the years. If we act decisively, this window of opportunity will allow us to make the digital world significantly safer."
The real threat landscape
The giants' concern is not unfounded — data confirms rapid escalation. TeamT5 analysts have recorded that Chinese hacker groups have doubled the number of attacks after switching to open models like DeepSeek. Moreover, they chose them not for their power, but for their low cost.
Even more telling figures were presented by Anthropic. After analyzing 832 blocked accounts, specialists found that the share of malicious actors with medium and high risk levels grew from 33% to 56% over the year. AI now independently handles privilege escalation and lateral network movement — tasks that previously required serious technical skills. The level of an attack is increasingly less dependent on the attacker's expertise.
South Korean researchers at Genians have found that North Korea's intelligence unit Kimsuky is actively testing local AI tools. Meanwhile, TRM Labs rated the use of AI for criminal purposes at 54 points out of 100 for this year, whereas in 2024 the figure was only 28. In the first half of 2026, 201 hacks of crypto projects were recorded — compared to 83 a year earlier.
My analysis: We are witnessing a turning point. AI cyberattacks have ceased to be an exotic phenomenon and have become an industry with a low entry barrier. The crypto sector, with its high liquidity and anonymity, remains the primary target — the nearly 2.5-fold increase in the number of hacks confirms this. Investors and project operators should view cybersecurity not as an expense item, but as a critical survival factor in the coming years.