This week's cybersecurity news is a ransomware report that reads like an escalation ladder. Semperis' 2026 Ransomware Risk Report — built with Censuswide from 1,200 IT and security professionals across 9 countries, every one of them hit by ransomware in the past year — finds that attackers weaponized AI in 84% of attacks. And the numbers keep getting darker from there: 87% of attacks targeted identity systems, 44% targeted the victims' own AI systems, and 40% included threats against employees and their families.
Released around October 14, 2026, the report surveyed 1,200 IT and security professionals in the US, UK, France, Germany, Italy, Spain, Singapore, Canada, and Australia, spanning 8 industry sectors. All respondents' organizations had been hit by ransomware in the preceding 12 months. The key figures:
Sources: Semperis 2026 Ransomware Risk Report; CBS News (Inglis interview); policy-wire.com; inview.org.uk; thearabianpost.com — October 2026.
Semperis CEO Mickey Bresman describes a three-stage escalation: first came encryption (pay or lose your data), then double extortion (encryption plus threats to leak stolen data), and now personal and family pressure. When 40% of attacks include psychological or physical threats against employees and their families — with 53% of intimidated victims saying attackers followed through, 37% reporting threats against relatives or friends, and 34% of victims or families being swatted — the ransom demand has moved beyond the server room and into the home.
Former US National Cyber Director Chris Inglis, now a Semperis strategic adviser, put it bluntly: "I think we're already seeing autonomous attacks." AI lets low-skill actors run sophisticated breaches — automating reconnaissance, crafting convincing lures, adapting to defenses in real time. And in a bitter irony, 44% of organizations reported attacks specifically targeting their own internal AI systems: the models and agents deployed to help the business have become attack surfaces of their own. "Holding you at physical risk is the latest phenomenon because we've covered most of the other risks," Inglis added — explaining why the 27% who sought psychological or medical treatment, the 24% who hired personal security, and the 14% who relocated or changed jobs are part of this story too.
BreachHunter — Ingenious AI's breach-intelligence platform — tracks the ransomware economy: which gangs are active, what they steal, and how attacks evolve. Get early access on WhatsApp.
1. What share of ransomware victims in the Semperis 2026 report said attackers weaponized AI?
2. What share of organizations faced attacks targeting their OWN AI systems?
3. What did the report say was targeted in 87% of attacks?
4. What did Chris Inglis call "the latest phenomenon" in ransomware tactics?
5. What did Inglis mean by "autonomous attacks"?
The Semperis 2026 Ransomware Risk Report, conducted with Censuswide and released around October 14, 2026, surveyed 1,200 IT and security professionals across 9 countries (US, UK, France, Germany, Italy, Spain, Singapore, Canada, Australia) and 8 industry sectors. Every respondent's organization had been hit by ransomware in the preceding 12 months. Its headline finding: 84% of victims said attackers weaponized AI in the attack.
According to the report, AI is being weaponized in 84% of ransomware attacks, letting lower-skilled actors run sophisticated breaches. In a darker turn, 44% of organizations experienced attacks specifically targeting their own internal AI systems — the very tools they deployed to help the business are being turned into attack surfaces.
Identity systems were targeted in 87% of the attacks surveyed. Semperis specializes in identity resilience, and the report's logic is straightforward: whoever controls identity controls access to everything. Compromising identity infrastructure turns every credential, session, and permission into a weapon.
The report found 40% of attacks included psychological or physical threats against employees and their families. Among those reporting intimidation, 53% said attackers followed through; 37% reported threats against relatives or friends; 34% of victims or their families were swatted; 27% sought psychological or medical treatment; 24% hired personal security; 14% relocated or changed jobs. Former US National Cyber Director Chris Inglis called physical risk 'the latest phenomenon because we've covered most of the other risks.'
Chris Inglis, now a Semperis strategic adviser, said: 'I think we're already seeing autonomous attacks' — AI systems that carry out attack stages with minimal human involvement. Combined with the report's finding that 59% of victims paid a ransom at least once and 69% suffered multiple successful breaches in a year, the picture is of faster, cheaper, and more relentless ransomware operations.