Collection of cybersecurity reports, surveys, research papers, market studies, and industry intelligence.
726 articles
A public-source review of Myanmar-linked cyber records from 2016 to 2026 finds recurring risks to trusted channels, identities, and institutional data, while stressing that selected reports cannot establish national incident prevalence or confirm every attribution.
A Huntress-commissioned survey of 461 U.S. professionals in regulated industries found 48% reported a major cyberattack at their organization, while 93% expect one within a year. Respondents also reported preparedness and security staffing gaps.
Huntress outlines cyber insurance trends, including tighter coverage, rising regulatory demands, data-theft extortion and BEC claims, and concern about AI-enabled attacks. It advises businesses to review policy terms and strengthen identity and email security.
Acronis’s global update, based on data from more than one million endpoints, covers an EY client-data breach and July 2026 malware and URL detection trends.
Kaspersky’s Q2 2026 report says malicious objects were blocked on 19.15% of ICS computers, the lowest share since 2022. Ransomware and email threats rose, while internet and removable-media threats declined.
Wiz’s 2026 report says contextual analysis removes more than half of high-priority cloud findings, while exposed access paths and privilege can create greater risk than isolated software flaws.
Kaspersky’s Q2 2026 analysis reports rising vulnerability disclosures, more exploits published before patches, active exploitation of Linux privilege-escalation flaws, and growing vulnerabilities in AI tools.
CYFIRMA’s Trust Cascade report compares four cases linking SaaS delegation, identity compromise and AI-agent risks. It prioritizes task-scoped authorization while warning that identity controls alone cannot stop prompt injection.
Cofense examines fake software-installation lures, Browser-in-the-Browser phishing, ClickFix, and device-code phishing, showing how attackers exploit trusted browser interactions to steal credentials or deliver malware without exploiting browser vulnerabilities.
CISA’s simulated tests found that a government organization failed to detect and respond effectively, while a water organization detected intrusions and quarantined systems. Both had cloud-security and token-management weaknesses.
Unit 42 analyzed 405 AI-related malware samples and found about 97% only in research or testing environments. Twelve appeared in protected endpoint telemetry, and Palo Alto Networks reported that its defenses detected and blocked them.
Mexico’s cybersecurity plan proposes new institutions, legislation, and regional cooperation. Insikt Group’s assessment identifies ransomware, state-linked activity, financial malware, data breaches, hacktivism, and cyber-enabled organized crime as key risks.
AhnLab’s July 2026 report says manufacturing had the most reported ransomware incidents, Gentlemen led group rankings, and established and newly emerged groups claimed attacks worldwide.
Truffle Security reports that more than 9,300 publicly exposed AWS keys remained active. Findings include company-linked and root keys, AdministratorAccess credentials, and 8,482 exposures from Hugging Face; researchers recommend revoking or rotating exposed keys.
StepSecurity says it tracked 56 confirmed malicious supply-chain compromises across trusted packages, GitHub Actions and IDE extensions in 12 months, with incidents rising sharply from March 2026. Its report highlights self-propagating worms and CI/CD credential theft.
Aikido benchmarked 10 AI models against 32 recent vulnerabilities, running each three times. DeepSeek V4 Pro found 28 across pooled runs; results also highlight model inconsistency, false leads, and cost trade-offs.
Sophos reviewed 38 AI-related MDR cases from July 2025 to June 2026 and found most involved attackers abusing AI brands, especially through fake installers, to deliver malware. Evidence of AI-assisted attack tooling was limited and often circumstantial.
Rapid7’s Q2 2026 analysis reports a sharp rise in high- and critical-severity vulnerability disclosures and says AI-assisted attacks are shortening response windows. It urges defenders to prioritize exposure over CVSS scores and monthly patch cycles.
Picus Security’s sponsored summary of Blue Report 2026 says more than 338 million simulations found Mimikatz credential-dumping prevention varied by method: 94% for LSASS memory access, 17% for alternate-memory access and 3% for LSA Secrets.
Recorded Future reports that PurpleDelta, a North Korean IT worker network, used fabricated personas and AI tools to apply to more than 1,100 companies and gain employment at at least ten organizations, posing insider and data risks.
Acronis analyzed about 545,000 patch applications across SMB Windows fleets from January to May 2026, finding that 11.2% needed closer review and that warning rates varied sharply by product.
The report examines attacks on developer accounts, packages, build pipelines and extensions, highlighting self-propagating worms and industrialized campaigns. It recommends controls including pinning, provenance, deliberate updates and dependency inventories.
Resecurity says a 152.5-GiB archive tied to the March 2026 LiteLLM/Trivy campaign contains 415,427 CI secret-capture files across 2,038 repositories; its sample was dominated by GitHub Actions tokens. Attribution draws partly on published reporting.
CYFIRMA profiles Russia-linked Turla’s long-running espionage and reports suspected STOCKSTAY use against government and defense entities in Ukraine and Italy, plus researchers’ report of a July 2026 intrusion campaign targeting French organizations.
