Section
Security & Defence
Toggle SidebarIAIThe Impact of AIEvidence-first reportingLanguageENRead in your language
87 reports
- Can two defences stop hidden backdoors in fine-tuned AI? Lab attacks fell sharply—but real compromised models were not testedSecurity & Defence
- Did AI break RSA? It accelerated two attacks on legacy 1024-bit settings—not modern 2048-bit encryptionSecurity & Defence
- Do India’s new AI audit guidelines create binding rules? They set a four-tier framework—but legal force is unclearSecurity & Defence
- What will Japan's emergency cyber package change—and where does AI fit?Security & Defence
- What should people do before giving an AI agent account access? Singapore set out five safety stepsSecurity & Defence
- Who secures a self-hosted open-weight AI model? Australia says the operator inherits the provider's jobSecurity & Defence
- Can an AI dispatcher respect fleet capacity? Hard safeguards rewrote fewer actions—but did not clearly finish more missionsSecurity & Defence
- Did an AI pipeline find 948 Linux paths to root? OpenAI reports exploits, patches and a validation warningSecurity & Defence
- What can GPT-6 Cyber do—and who can use it?Security & Defence
- Can CrowdStrike stop a rogue AI agent on the endpoint today?Security & Defence
- Has AI created a new class of UK cyber attack? The NCSC says not yetSecurity & Defence
- What did Saab's in-house AI do in a Gripen E—and what remains untested?Security & Defence
- Chainalysis says names and emails were accessed—what does ‘agentic’ prove?Security & Defence
- Does analyst feedback make a lightweight cyber detector field-ready?Security & Defence
- Northflank says rotate every stored secret now—what should customers do first?Security & Defence
- A coding agent can look normal while a backdoored extension steals credentialsSecurity & Defence
- A malicious Claude Code extension planted an SSH backdoor—updating alone is not enoughSecurity & Defence
- An open model built Firefox file-theft exploits across six releasesSecurity & Defence
- A smart-grid detector topped 0.98 AUROC in simulation—real utilities remain untestedSecurity & Defence
- Europe put €42m behind AI cyber-defence projects—but results do not exist yetSecurity & Defence
- GLM‑5.3 completed 3 of 13 simulated cyber operations—how much does that show?Security & Defence
- Radiation-dose AI matched 12 of 13 test samples—but was not tested in an emergencySecurity & Defence
- An internal Claude model built five exploit chains; the disclosed flaws were patched before publicationSecurity & Defence
- Can AI spot a lie? Strong synthetic scores collapsed on real military interviewsSecurity & Defence
- Can every AI agent have a revocable identity? IBM has shipped part of the control stackSecurity & Defence
- How did 155 AI responses expose another customer’s data? A patched cache boundary failedSecurity & Defence
- Australia says AI is speeding cyber threats—but its incident count does not isolate AISecurity & Defence
- What will the UK's £300mn AI Biosecurity Lab actually deliver?Security & Defence
- Why are cyber agencies urging emergency Fortinet updates?Security & Defence
- Why are five cyber agencies treating AI tokens as an operational risk?Security & Defence
- AI agents retrieved the right memory—and still used a version they should notSecurity & Defence
- An AI agent can act unsafely when one true fact is missing—even if nothing it sees is falseSecurity & Defence
- Does continuous AI code scanning show when a vulnerability is really fixed?Security & Defence
- NVIDIA's OpenShell fences in AI agents—but it cannot prove they are safeSecurity & Defence
- Could a better benchmark score hide a poisoned training dataset?Security & Defence
- Could a privacy audit falsely accuse a correct AI system?Security & Defence
- Could an AI agent prove each action keeps its contract before it acts?Security & Defence
- Could your own voice trigger a poisoned AI assistant?Security & Defence
- Malicious subagents repeatedly talked their way into off-limits work—what does the benchmark show?Security & Defence
- PHP2Shell attack attempts began before disclosure—what must operators do now?Security & Defence
- What must be tested before AI agents defend real networks?Security & Defence
- AI agents probed 1,200 sites—what should defenders change first?Security & Defence
- Microsoft says its new controls can contain risky AI agents—what do they actually cover?Security & Defence
- US agencies set ten AI-agent incident practices—what changes for operators?Security & Defence
- Washington is removing DeepSeek from civilian federal IT—how far does the order reach?Security & Defence
- A malware classifier hit 100% on a benchmark—not on live attacksSecurity & Defence
- One error bit was enough to recover encrypted-computing keysSecurity & Defence
- AI-assisted attacks are hitting Indian finance at machine speed. What must firms do now?Security & Defence
- Can a malicious instruction travel from one AI agent to another?Security & Defence
- Can a simple projection reduce what RAG systems leak?Security & Defence
- Can formatting change an AI safety guard's decision?Security & Defence
- Can generative AI hide wireless data from eavesdroppers?Security & Defence
- Does an AI agent's safety guard change with language?Security & Defence
- How can an approved AI tool read files it never declared?Security & Defence
- How often did prompt injection divert coding agents?Security & Defence
- What did the agent-security recreation actually prove?Security & Defence
- When an AI agent hits a blocked website, should it keep trying?Security & Defence
- Can a high score prove an Android-malware AI is reliable?Security & Defence
- Can encoded instructions turn AI-agent skills into security exploits?Security & Defence
- Can AI help security teams test less-visible attack paths?Security & Defence
- Can an AI agent finish the job and still harm the workspace?Security & Defence
- Will Britain require serious AI incidents to be reported?Security & Defence
- Do specialist tools amplify AI biosecurity risk?Security & Defence
- Can AI score software vulnerabilities before analysts do?Security & Defence
- Can federated AI detect IoT botnets without pooling traffic?Security & Defence
- Can Anthropic’s free AI scanner help open-source maintainers?Security & Defence
- Can federated health AI resist poisoned model updates?Security & Defence
- Can reduced AI safeguards give cyber defenders an advantage?Security & Defence
- Do low error rates prove biometric templates are secure?Security & Defence
- AI in Ukraine’s defence: what it changes for peopleSecurity & Defence
- Can AI detect attack types it never saw in training?Security & Defence
- Can phishing models learn without pooling URLs?Security & Defence
- Can clouds share phishing signals without sharing raw data?Security & Defence
- Can page content expose malicious websites?Security & Defence
- Are cyber budgets rising faster than AI safeguards?Security & Defence
- Will Apple’s new consent controls make AI agents safer on Macs?Security & Defence
- How did attackers try to extract hidden reasoning from OpenAI models?Security & Defence
- Is AI changing cyberattacks—or speeding up familiar tactics?Security & Defence
- Who holds AI agents accountable when they reach government systems?Security & Defence
- Why are China-aligned phishers impersonating AI-policy insiders?Security & Defence
- Can companies control what AI agents can access? Gartner finds a governance gapSecurity & Defence
- What must Australian agencies change after the new AI-enabled cyber direction?Security & Defence
- What must UK AI researchers do after MI5's CGTRI espionage alert?Security & Defence
- Why do promising AI pilots struggle to reach UK security teams?Security & Defence
- Shared files carried a simulated attack between AI assistants; real-world spread is unprovenSecurity & Defence
- Palo Alto Networks launches continuous AI-led exposure testingSecurity & Defence
- AI is scaling cyber work for attackers and defenders—and evidence quality mattersSecurity & Defence