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.
Summit to Focus on Open-Source, AI Governance and DevelopmentThe historic presidential Élysée Palace in Central Paris will host world leaders, tech CEOs and researchers for the French AI Action Summit, a two-day event that will commence on Monday. U.S. Vice President JD Vance, OpenAI CEO Sam Altman and Google's Sundar Pichai will be on hand.
Governments Are Skeptical of Chinese A1 Platform's Data Security ControlsCountries across Asia are racing to ban government officials, national agencies and critical infrastructure organizations from using Chinese artificial intelligence company DeepSeek's open-source chatbot application, citing data security and privacy risks.
The secret use of other people's generative AI platforms, wherein hijackers gain unauthorized access to an LLM while someone else foots the bill, is getting quicker and stealthier by the month.
Also: Addressing AI Vulnerabilities and Governance ChallengesDeepSeek, an advanced open-source AI model, is under scrutiny for its safety guardrails failing multiple security tests and a data leak that exposed user information and API keys. Sam Curry, CISO at Zscaler, discusses AI security, risk management and upcoming U.S. policy changes.
The technique, called nullifAI, allows the models to bypass Hugging Face’s protective measures against malicious AI models
Microsoft Edge 133 is now rolling out globally, and it ships with several improvements, including a new scareware blocker feature. In addition, Microsoft is updating the backend of the Downloads UI with performance improvements. [...]
The foundations for social engineering attacks – manipulating humans – might not have changed much over the years. It’s the vectors – how these techniques are deployed – that are evolving. And like most industries these days, AI is accelerating its evolution. This article explores how these changes are impacting business, and how cybersecurity leaders can respond
Also: Researchers Bypass GitHub Copilot's Protections, Deloitte Pays $5M for BreachThis week: A hacker claims to have 20 million OpenAI logins, Sweden clears ship in Baltic cable damage, researchers find ways to bypass GitHub Copilot's protections, Netgear patches router flaws, undetectable Mac backdoor raises alarms, Spain nabs hacker, and Deloitte pays $5M for RIBridges breach.
Cybereason co-founders launch their second act with a security startup focused offering a platform that uses agentic AI to offload repetitive tasks commonly performed by security analysts.
OpenAI's latest tech can reason better than its previous models could, but not well enough to ferret out careful social engineering.
New mobile apps from the Chinese artificial intelligence (AI) company DeepSeek have remained among the top three "free" downloads for Apple and Google devices since their debut on Jan. 25, 2025. But experts caution that many of DeepSeek's design choices -- such as using hard-coded encryption keys, and sending unencrypted user and device data to Chinese companies -- introduce a number of glaring security and privacy risks.
Riding the wave of notoriety from the Chinese company's R1 AT chatbot, attackers are spinning up lookalike sites for different malicious use cases.
AI-Powered Social Engineering and Deepfake Threats in 2025Security researchers predict threat actors will use artificial intelligence and large language models to enhance phishing attacks and create convincing fake personas, while defensive AI enters a new phase of semiautonomous operations.
Cisco has released updates to address two critical security flaws Identity Services Engine (ISE) that could allow remote attackers to execute arbitrary commands and elevate privileges on susceptible devices
Analyst Allie Mellen on Open-Source AI Adoption, Vendor Considerations, Data RisksAI adoption is accelerating across security operations, but DeepSeek has introduced security, privacy, and geopolitical risks that organizations should carefully assess. Forrester's Allie Mellen shares advice on AI adoption by cybersecurity, third-party risks and data protection.
Could Open-Source AI Redefine the Future? Here's What Experts SayOpen-source AI is shaking up the industry, challenging traditional large and small language models and raising new security concerns. With DeepSeek-R1 leading the charge, experts weigh in on the risks, rewards and the future of AI monetization. Is the future of AI open - or are we headed for new challenges?
Fraud groups are using AI and deepfake technology to scale up their operations to create fake identities and execute fraud campaigns.
The security startup's autonomous security remediation platform uses off-the-shelf large language models (LLMs) to analyze security alerts and apply the fixes.
'Frontier AI Framework' Identifies Risk Categories, Action PlanMeta has set new limits on the release of its advanced artificial intelligence models, establishing a framework detailing the criteria for restricting systems deemed too dangerous for public release. Meta's Frontier AI Framework identifies two risk categories: high and critical.
OneTrust's Ojas Rege Details Act Requirements, AI Governance ChallengesThe first set of rules banning high-risk AI systems under the European Union AI Act went into effect on Sunday. Starting this week, companies are now barred from deploying AI-driven emotion recognition in the workplace and schools. OneTrust's Ojas Rege discusses the implications.