惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Microsoft Azure Blog
Microsoft Azure Blog
WordPress大学
WordPress大学
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
The Cloudflare Blog
U
Unit 42
D
Docker
Hugging Face - Blog
Hugging Face - Blog
博客园 - 聂微东
Recent Announcements
Recent Announcements
GbyAI
GbyAI
T
The Blog of Author Tim Ferriss
Last Week in AI
Last Week in AI
V
Visual Studio Blog
I
InfoQ
Google DeepMind News
Google DeepMind News
小众软件
小众软件
L
LangChain Blog
C
Check Point Blog
宝玉的分享
宝玉的分享
Martin Fowler
Martin Fowler
酷 壳 – CoolShell
酷 壳 – CoolShell
博客园 - 【当耐特】
J
Java Code Geeks
罗磊的独立博客

The Register - Security: Research

Novel Blue Moon kit targeting Chrome and Windows reflects new reality of AI-driven exploits Extortion crews have their eyes on high-value AI data, Google warns Researcher shows how Claude Code can be tricked simply by asking it to summarize a website Copilot tricked into telling reseachers how to hack itself Akira ransomware scum blocked victim How the famed USENIX Security conf is managing a flood of papers in the AI era www.theregister.com Self-destructing Mistic backdoor linked to access broker selling corporate footholds to ransomware gangs PRC-linked spies hid inside medical and military networks for more than a year, snooping through Gmail and stealing data Nobody needs Mythos or 0-days to build a chaos-causing computer worm – free open source models work just fine ChatGPT blindly trusts browser content, turning the page into a payload Russia-linked threat group put ChatGPT to work from lure to payload Kids can bypass some age checks with a drawn-on mustache What type of 'C2 on a sleep cycle' do they leave behind? Novel Chinese spy group found in critical networks in Poland, Asia ORNL builds more sensitive GPS interference detector Researchers find sabotage malware that may predate Stuxnet Vibe coding upstart Lovable denies data leak, cites 'intentional behavior,' then throws HackerOne under the bus Anthropic, Google, Microsoft paid AI bug bounties – quietly Security reserchers tricked Apple Intelligence into cursing Don't open that WhatsApp message, Microsoft warns Security boffins harvest bumper crop of API keys from web Lightning-fast exploits mean patch fast, says Cisco Talos AI agents are 'gullible' and easy to turn into your minions Smooth criminals talking their way into cloud environments, Google says Snoops plant info-stealing malware on iPhones, Google warns Cybercrime up 245% since the start of the Iran war Rogue AI agents can work together to hack systems Fake applicants are sending security-killing malware AI agent hacked McKinsey chatbot for read-write access Kaspersky: No signs Coruna iPhone exploit kit made by US
Overconfidence is the new zero-day as teams stumble throu...
2025-11-17 · via The Register - Security: Research

Teams that think they're ready for a major cyber incident are scoring barely 22 percent accuracy and taking more than a day to contain simulated attacks, according to new data out Monday.

Immersive's latest Cyber Workforce Benchmark, which draws on 1.8 million exercises from the Immersive One platform and a survey of 500 cybersecurity leaders, paints a picture of an industry that has become more confident but no more capable. Immersive says 94 percent of organizations believe they can "effectively detect, respond to, and recover from a major incident," yet real-world performance in controlled drills has remained stubbornly flat. 

According to the report, resilience scores haven't improved since 2023, with the median response time to complete critical cyber threat intelligence labs still coming in at 17 days – despite what Immersive describes as "record investment" and growing pressure from boards and cyber insurance carriers. 

James Hadley, Immersive founder and chief innovation officer, argues that organizations are failing not for lack of effort, but because they are training for the wrong fights. "Readiness isn't a box to tick, it's a skill that's earned under pressure," he says in the report. "Organizations aren't failing to practice; they're failing to practice the right things." 

Across the company's crisis-simulation drills, which involved 187 professionals in 11 global exercises, performance was consistently poor. Participants achieved just 22 percent accuracy, averaged 60 percent confidence, and took 29 hours to contain an infection, a combination the report describes as evidence that "when tested under pressure, most teams didn't fail for lack of knowledge, they failed for lack of practiced coordination." 

The data also shows no improvement in the industry's basic readiness metrics. Immersive says more than 60 percent of sectors actually experienced slower response times year-over-year, and that confidence scores for "OK," "Good," and "Great" answers averaged the same (around 42.5 percent), suggesting teams cannot accurately judge their own performance despite expressing strong self-belief.

Much of the stagnation, the report argues, comes from practicing outdated threat scenarios. Immersive found that 60 percent of all training activity still focuses on vulnerabilities more than two years old, leaving teams "over-prepared for yesterday's threats" while new attacker techniques continue to evolve. Fundamental-level labs remain the most common exercises at 36 percent of usage, which the company says limits progression to intermediate and advanced readiness.

Another systemic issue is participation. Only 41 percent of organizations include non-technical roles such as legal, HR, communications, or senior executives in their cyber-response simulations. This is despite 90 percent of respondents believing their cross-functional communication during an incident is effective. Immersive's data shows the opposite: when business functions aren't rehearsed under pressure, collaboration falters and response times worsen. 

Industry habits also contribute to the readiness illusion. Immersive reports that organizations overwhelmingly rely on training completion rates to measure preparedness even though completion "is not competence." Only 46 percent use resilience scores, and only 42 percent measure the number of simulations conducted, creating what the report calls "false metrics" that mask real-world capability gaps. 

The report highlights a widening adaptability problem as well. Experienced practitioners perform strongly on familiar threats (roughly 80 percent accuracy in classic incident-response labs) but fall behind when faced with AI-enabled or novel attacks. Senior participation in AI-scenario labs dropped 14 percent year-over-year, while non-technical managers increased participation by 41 percent.

As Immersive puts it: "Experience teaches what to do next – until the next thing has never happened before."

Training completion itself remains inconsistent. The report notes an average completion rate of 81 percent, meaning nearly one in five participants do not finish the exercises they start.

Hadley argues the industry must shift from confidence built on assumptions to readiness grounded in evidence. "True resilience comes from continuously proving and improving readiness across every level of the business, so when a real crisis hits, your confidence is backed by evidence, not assumption."

"Experience teaches what to do next, until the next thing has never happened before," added Hadley. "Even the most seasoned teams must evolve as fast as the threats they face." ®