Criminal IP Brings AITEM to Tokyo as Japan Ransomware Hits Record

A booth in Tokyo during a hard year for Japanese security

Criminal IP, the threat intelligence platform run by AI SPERA, will exhibit at Security Days Fall 2026 Tokyo. The event runs October 21–23 at the TAKANAWA GATEWAY Convention Center, and the company will be at booth B20.21 It plans live demonstrations of two products: Criminal IP Threat Intelligence (Criminal IP TI) and Criminal IP AITEM, which stands for AI-Powered Threat Exposure Management. The demos will focus on threat investigation and exposure management.

On its own, this is a routine trade-show announcement. The setting gives it more weight. Japan is having one of its worst stretches of ransomware on record. Several large companies have spent months rebuilding, regulators are moving from advice to rules, and the time between a vulnerability becoming public and attackers using it has shrunk sharply. The real question for buyers is whether products like AITEM fix the failures behind Japan's biggest breaches, or just give those failures a new label.

What Criminal IP says it will demonstrate

The company's announcement lists six demo scenarios:

• investigating threat data tied to IP addresses, domains and URLs
• analyzing malicious infrastructure, vulnerabilities and threat actors
• finding IT assets and services exposed to the internet
• using AI-assisted analysis to rank exposure risks
• spotting leaked data on the dark web
• assessing risk from business partners and affiliates

21 The stated pitch is that threat intelligence gives context to exposure findings, so teams can go from knowing what assets they have to deciding what to do about them.21

AI SPERA CEO Byungtak Kang put it this way: knowing which assets are exposed is not enough unless teams also understand how those exposures connect to actual threats. He said combining intelligence with AI-assisted exposure management helps organizations focus on the risks that matter.4 Coverage that rewrote the release kept that framing and stressed prioritization for enterprises.24

The company's own event page puts the emphasis somewhere slightly different. It describes the Tokyo demos mainly as continuous monitoring of internet-facing domains, IP addresses, ports and services to find unknown assets. Combining asset visibility with threat intelligence comes second. Our reading is that the established product here is attack surface management (ASM), meaning tools that map what an organization exposes to the internet. AITEM is a framework being built on top of it.

AITEM: a product, or a plan for one

AITEM was announced a week earlier, alongside Criminal IP's appearance at GovWare 2026 in Singapore. The release described it as the company's "vision" for where attack surface management goes next, not as a finished, separately priced product.2 According to that release, AITEM goes beyond a list of external assets. It pulls in open-source intelligence, dark web data, internal infrastructure, Shadow AI (AI tools staff use without approval), leaked data and newly disclosed vulnerabilities.1 It is also meant to feed prioritized findings into alerts, tickets and workflow actions sent to the right teams.9

The company's reasoning is that AI has made attacking cheaper. Automated scanning, published proof-of-concept exploits and AI-assisted vulnerability discovery let attackers find exposed assets faster. In that view, the defender's main problem is how fast they can respond, not whether they can see.8 Criminal IP links this to wider industry trends. It notes that agentic AI, AI-driven security operations centers and Shadow AI were major themes at RSAC 2026.1

One independent threat-tracking service rated the AITEM news as low severity. It pointed out that the announcement describes a product, not a vulnerability, and suggested that organizations interested in ASM evaluate AITEM "or similar" tools.6 That is the right attitude. Nearly every vendor now uses the language of exposure management, and the claims in the announcement have not been independently tested.

Japan's ransomware numbers

The Tokyo show itself puts ransomware front and center. Organizer Nano Opt Media markets it as a free source of incident trends and defenses against ransomware and targeted attacks. Cumulative session attendance at the Tokyo venue is around 16,000.

The latest figures explain why attendees will care. Japan's National Police Agency recorded 123 ransomware cases in the first half of 2026, a record. Suspicious access attempts at police monitoring points also hit a record average of 13,687 per IP address per day.19 Further analysis of the police data found:

• 31 of the 123 cases hit major companies
• more than half took over a month to recover from
• nine shut down business operations entirely
• about 60% caused losses above 10 million yen

18

Outside researchers agree on the trend but count differently. Forescout found that ransomware claims against Japanese organizations rose 39% year over year in early 2026, faster than the 25% global increase. Japan moved up to 14th among the most-targeted countries. Forescout also now tracks 124 threat actors targeting Japan, 82% more than in 2024.14 Cisco Talos, as cited by CybelAngel, counted 134 incidents in Japan in 2025. Qilin was the most active group, responsible for 16.4% of cases, and small and mid-sized companies made up 57% of victims.15

The gap between these numbers comes from method. Forescout counts public leak-site claims. The police count reported cases. A national survey found that at least 507 Japanese companies had been hit by ransomware at some point, and 44% of them paid.14 All of the counts point the same way, and the public figures almost certainly undercount the real total.

How the biggest breaches started

The two most-studied incidents, at Asahi and Askul, show where exposure management would apply. A comparison by a Japanese law firm traced Askul's breach to a contractor's admin account that did not have multi-factor authentication. Asahi's attackers came in through a VPN appliance at one of its sites.17 SecurityWeek reported that Askul's attackers used stolen credentials, explored the network, moved between systems, turned off security tools and deleted backups before encrypting files.11 Asahi's attackers reportedly spent about ten days inside the network, mostly outside business hours, before launching ransomware.18

The damage went well beyond the two companies. Askul's outage took down the online store of Ryohin Keikaku, which runs Muji, because it relies on Askul for deliveries.16 The RansomHouse group claimed it took 1.1 terabytes of data from Askul.20

Reports of the scale don't match:

• Askul records: SecurityWeek reported more than 700,000 compromised records.11 The law-firm analysis puts it at about 740,000.17
• Asahi records: About 1.91 million people were initially feared affected, but only about 115,000 were confirmed.17 Some outlets still cite the larger number.15
• Askul recovery: Estimates range from six weeks to restart corporate orders15 to three and a half months for full recovery.12 Core services were restored on February 13, 2026.17

The early, alarming figures from these incidents often got revised later, so initial breach counts deserve some caution.

The common thread is that both companies were breached through ordinary weaknesses, not exotic techniques: a weakly protected account, an exposed remote-access device, and a supplier connection. These are the exposed assets, leaked credentials and partner risks that Criminal IP plans to demo.21 CybelAngel's advice to Japanese teams is close to the same list: watch for exposed credentials continuously, because Qilin buys initial access from Telegram and breach forums, and map how far a supplier's failure would spread.15 The match with Criminal IP's demos is real. That doesn't prove any single tool would have prevented these attacks.

Less time to patch

Vulnerability response is the area where AI-assisted prioritization could matter most. Yokohama National University professor Katsunari Yoshioka is cited as finding that the average time from public disclosure of a vulnerability to its exploitation fell from 141 days in 2024 to 35 days by the end of August 2026.12 Forescout reported that the number of published vulnerabilities rose 51% year over year in the first half of 2026.14 CybelAngel describes a China-linked group, Storm-1175, that uses zero-day exploits and deploys Medusa ransomware in under 24 hours.15

This week's headlines added more. They included critical fixes in SAP's October 2026 security updates, a critical Ghostscript flaw that can be triggered by malicious PDFs, a weakness in npm's Trusted Publishing feature that put more than 1,500 packages at risk of takeover, and Kimsuky delivering Funhouse ransomware through booby-trapped résumé documents. With that many disclosures and that little time, teams can't patch everything first. Deciding what to fix first, based on whether something is actually exposed and under attack, is the strongest argument for this whole product category.

Regulators raise the bar

Tokyo is also tightening rules. Common Cybersecurity Standards for Critical Infrastructure took effect on October 1, the same day a new public-private cyber cooperation council started work. The Financial Services Agency then told banks to review their defenses. Coverage says the trade ministry (METI) has asked about 1,000 industry groups to check their systems.18 Japanese manufacturers with large supplier networks should expect more detailed security questionnaires from insurers and big customers.18 That gives vendors offering partner-risk scoring and external visibility, as Criminal IP does, a real compliance-driven opportunity.

Our assessment

Criminal IP's Tokyo appearance makes sense commercially. Its messaging also lines up with what the incident record shows: Japan's costliest recent breaches started with exposed access points, leaked credentials and weak links to suppliers.1715 The company is also expanding internationally. It appointed a Middle East and Africa distributor in September and launched AITEM at GovWare weeks before Tokyo.212

Buyers should still keep the vision apart from what has been shown working. AITEM was introduced as a direction for the company, and the Tokyo demos look built mainly on Criminal IP's existing scanning and intelligence products.2 Japan's own after-action reviews stress basics such as multi-factor authentication, immutable backups and network segmentation.17 Exposure management can help teams find their gaps faster. It does not close them by itself. The test for Criminal IP in Tokyo is whether its live demos show defenders acting faster, rather than just seeing more.