Explore the intersection of AI and cybersecurity. Stay informed on AI-driven security trends, tools, and threats in the ever-evolving digital landscape.
Search across headline titles and summaries.
Background for this topic.
Artificial intelligence (AI) describes computer systems that perform tasks such as recognizing patterns, making predictions, understanding language, or generating content. In security reporting, the term commonly includes machine-learning models used for detection and analysis, as well as generative AI applications that produce text, code, images, or other outputs.
AI can help analyze security telemetry, prioritize vulnerabilities, and support investigations, but its outputs can be wrong or manipulated. Important attack surfaces include prompt injection that steers an application into unintended actions, sensitive data being exposed through prompts or model outputs, and excessive permissions granted to AI systems that use external tools. Models can also be degraded by poisoned training data or evaded with carefully crafted inputs. Practitioners should protect training and operational data, limit model access and tool permissions, test for adversarial behavior, and require appropriate human validation before high-impact decisions.
Researchers Show How AI Image Downscaling Can be an Attack VectorResearchers discovered a method to embed invisible prompt injections that are activated during AI's processing of an image. When the model scales down these images, the hidden malicious instructions allow theft of data from popular image production systems.
The growing ecosystem of agents, chatbots, and machine credentials that outnumber human users by an order of magnitude is creating a poorly understood but potentially major security issue.
AI-assisted security reviews from Anthropic and others could help level up enterprise application security in the era of vibe coding.
Discover how AI is reshaping cybersecurity through our CEO, Eva Chen’s industry briefing series. Gain practical strategies, real-world insights, and a clear roadmap to secure your AI initiatives with confidence.
Because savvy terrorists always use public internet services to plan their mischief, right? Anthropic says it has scanned an undisclosed portion of conversations with its Claude AI model to catch concerning inquiries about nuclear weapons.…
CEO Amir Ben-Efraim: Acquisition Adds AI-Powered File Sanitization to Browser ToolsThrough its acquisition of Votiro, Menlo Security has embedded file-level sanitization and AI-powered detection directly into its enterprise browser stack. CEO Amir Ben-Efraim says the move helps prevent malware, data leaks and phishing risks at the browser level.
Copilot Falls for Prompt Injection Yet AgainMicrosoft quietly fixed a flaw that allowed users to instruct embedded artificial intelligence model Copilot not to log its access corporate files. "If you work at an organization that used Copilot prior to Aug 18, there is a very real chance that your audit log is incomplete."
Google’s Gemini-powered tools tripped up by image-scaling prompt injection Security researchers with Trail of Bits have found that Google Gemini CLI and other production AI systems can be deceived by image scaling attacks, a well-known adversarial challenge for machine learning systems.…
By using brief, plain clues in their prompts that are likely to influence the app to query older models, a user can downgrade ChatGPT for malicious ends.
NCC Group's David Brauchler III shared how foundational controls and threat modeling strategies can help secure agentic AI tools in ways traditional guardrails can't.
Commvault has released updates to address four security gaps that could be exploited to achieve remote code execution on susceptible instances
From hybrid cloud to AI, IT complexity and security risks are accelerating. Certified VMware pros bring clarity and resilience that keep teams ahead. See why CIOs are making certification a workforce strategy with VMUG. [...]
Trag's Developer-Centric Tools Help Aikido Slash Time to Market by 12 MonthsAikido Security acquired Trag, an AI-native code review startup, to bring repository-wide review capabilities to its platform. The acquisition accelerates delivery of new features, such as logic risk detection and English-language rule writing, aimed at beating legacy rivals.
DARPA's Kathleen Fisher discusses the AI Cyber Challenge at DEF CON 33, and the results that proved how automation can help patch vulnerabilities at scale.
One fetcher bot seen smacking a website with 39,000 requests per minute Cloud services giant Fastly has released a report claiming AI crawlers are putting a heavy load on the open web, slurping up sites at a rate that accounts for 80 percent of all AI bot traffic, with the remaining 20 percent used by AI fetchers. Bots and fetchers can hit websites hard, demanding data from a single site in thousands of requests per minute.…
Guardio reveals a new AI take on ClickFix dubbed “PromptFix”
OpenAI's CEO Sam Altman told reporters that GPT-6 is already in the works, and it'll not take as long as GPT-5. [...]
Cybercriminals are increasingly abusing the AI-powered Lovable website creation and hosting platform to generate phishing pages, malware-dropping portals, and various fraudulent websites. [...]
Move along, nothing to see here Amazon has quietly fixed a couple of security issues in its coding agent: Amazon Q Developer VS Code extension. Attackers could use these vulns to leak secrets, including API keys from a developer's machine, and run arbitrary code.…
Claude Models May Shut Down Harmful Chats in Some Edge CasesAnthropic introduced a safeguard to its Claude artificial intelligence platform that allows certain models to end conversations in cases of persistently harmful or abusive interactions. The company said it's doing so not to protect human users, but as a way to mitigate risks to the models.