Research examines attack methods, defenses, and vulnerabilities, helping security teams understand risks and improve protection.
Search across headline titles and summaries.
Background for this topic.
Research is the systematic study of technologies, systems, attack methods, vulnerabilities, and defensive techniques to establish evidence and produce new findings. In information security, it includes work such as discovering flaws in software or protocols, analyzing malware and attacker behavior, testing cryptographic designs, and evaluating security controls. News under this tag may describe a proof of concept, a measurement study, or a proposed technique rather than a confirmed real-world attack.
For practitioners, research can change how risks are prioritized and mitigated. A demonstrated vulnerability may require vulnerability-management teams to verify affected assets, apply fixes, or add compensating controls; responsible disclosure gives developers time to assess and remediate before technical details enable exploitation. Research involving live systems, personal data, or offensive tooling also raises privacy, authorization, dual-use, and ethical concerns. Sound findings should state their assumptions, scope, limitations, and reproducibility, since laboratory results do not automatically show that an attack is practical in every environment.
Eset: Lazarus Group Shares Backdoor With Newer Pyongyang Threat ActorA gang of North Korean hackers behind fake IT job recruitment scams now have access to a remote access Trojan favored by their more technically advanced counterparts tracked collectively as the Lazarus Group, say security researchers.
Agentforce Agentic AI Tool Was Exposed to Indirect Prompt Injection AttacksSalesforce has patched a vulnerability involving its Agentforce agentic artificial intelligence tool, discovered by researchers, that attackers could have exploited using an indirect prompt injection attack to steal sensitive customer data and leads being stored in the CRM system.
Government Agencies Targeted With Infostealers and Cryptomining MalwareA fake police alert is the social engineering cornerstone of an ongoing phishing campaign targeting Ukrainian government agencies, warn security researchers. They're tracking a surge in malicious SVG file attachments, in this case leading to information-stealing and cryptocurrency-mining malware.
OpenAI, Apollo Research Find Models Hide Misalignment; Training Cuts DeceptionFrontier artificial intelligence models are learning to hide their true intentions to pursue hidden agendas, said OpenAI and Apollo Research. Researchers say the risk of deception needs to be tackled now, especially as AI systems take on more complex, real-world responsibilities.
A new campaign has been observed impersonating Ukrainian government agencies in phishing attacks to deliver CountLoader, which is then used to drop Amatera Stealer and PureMiner
Security researchers say multiple threat groups, including Iran's Charming Kitten APT offshoot Subtle Snail, are deploying malware with code-signing certificates from the Houston-based company.
Researchers say tens of thousands of instances remain publicly reachable Security researchers have confirmed that threat actors have exploited the maximum-severity vulnerability affecting Fortra's GoAnywhere managed file transfer (MFT), and chastised the vendor for a lack of transparency.…
More fun with AI agents and their security holes A now-fixed flaw in Salesforce’s Agentforce could have allowed external attackers to steal sensitive customer data via prompt injection, according to security researchers who published a proof-of-concept attack on Thursday. They were aided by an expired trusted domain that they were able to buy for a measly five bucks.…
Cybersecurity researchers have discovered an updated version of a known Apple macOS malware called XCSSET that has been observed in limited attacks
Yet again researchers have uncovered an opportunity (dubbed "ForcedLeak" for indirect prompt injection against autonomous agents lacking sufficient security controls — but this time the risk involves PII, corporate secrets, physical location data, and so much more.
The security researchers who discovered the malicious npm package called it the “first malicious MCP in the wild”
Cybersecurity researchers have disclosed a critical flaw impacting Salesforce Agentforce, a platform for building artificial intelligence (AI) agents, that could allow attackers to potentially exfiltrate sensitive data from its customer relationship management (CRM) tool by means of an indirect prompt injection
The hackers are likely trying to collect data to feed the development of zero-day exploits, said Google researchers
Cybersecurity researchers have discovered two malicious Rust crates impersonating a legitimate library called fast_log to steal Solana and Ethereum wallet keys from source code
Trend™ Research analyzed source binaries from the latest activity from notorious LockBit ransomware with their 5.0 version that exhibits advanced obfuscation, anti-analysis techniques, and seamless cross-platform capabilities for Windows, Linux, and ESXi systems.
Researchers Uncover Covert Chinese Access to US Service Provider InfrastructureMandiant said it has tracked a Chinese-linked espionage campaign using BRICKSTORM malware to quietly embed within U.S. infrastructure and service providers for over a year, exploiting appliance-level blind spots to maintain persistence, evade detection and potentially develop zero-day exploits.
Researchers have tracked a Russian disinformation campaign against upcoming Moldovan elections, linking it to a previous campaign that began in 2022.
'More Research' Needed, Says GoogleGoogle DeepMind expanded its risk framework to cover scenarios where artificial intelligence models might manipulate people or resist shutdown, marking the company's most explicit warning yet about potential misalignment. DeepMind frames many of these scenarios as a malfunction.
Cybersecurity researchers have disclosed two security flaws in Wondershare RepairIt that exposed private user data and potentially exposed the system to artificial intelligence (AI) model tampering and supply chain risks
Cybersecurity researchers have disclosed details of a new malware family dubbed YiBackdoor that has been found to share "significant" source code overlaps with IcedID and Latrodectus