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.
Panelists Discuss Deepfake, Trust Frameworks, AI Skepticism, Venture Capital WoesFrom RSAC Conference 2025 in San Francisco, ISMG editors wrapped up coverage discussing the impact of U.S. government funding cutbacks, growing deepfake threats, trust challenges in AI adoption and venture capital pressures affecting the cybersecurity vendor market.
Rain Capital's Chenxi Wang Warns About AI's Emerging Role in CodingAI can significantly accelerate code generation - helping developers go from idea to implementation in minutes - but AI-generated code is frequently based on insecure or flawed examples found in public code repositories, explains Chenxi Wang, founder and general partner at Rain Capital.
NightDragon CEO Dave DeWalt Sees Cautious Optimism With Agentic AIAdvances in scalable AI and agentic technologies are creating a long-awaited shift in the defender-offender dynamic. With autonomy and agentic capabilities entering production, CISOs may soon deploy 100,000 autonomous agents instead of hiring more staff, said Dave DeWalt, CEO of NightDragon.
Google NotebookLM, which is a research and note-taking AI tool, is getting upgraded to Gemini 2.5 Flash. [...]
Accountability Is Key as Enterprises Adopt AI at Scale, Says Saviynt's Jim RouthAI governance must balance innovation with security, making it vital that organizations adopt flexible, consensus-driven approach to ensure responsible AI deployment while addressing risks such as data exposure and software resilience, said Jim Routh, chief trust officer at Saviynt.
Estonia's Cyber Diplomat Tanel Sepp Urges Trust-First AI StrategyEstonia's Ambassador at Large for Cyber Diplomacy Tanel Sepp emphasizes the need for global cooperation and trust in AI adoption in digital economies. He discusses Estonia's whole-of-society model, the importance of regional partnerships and why digital sovereignty still lacks clear definition.
Tanium's Dan Streetman on Why Defenders Need to Optimize ToolingGood AI defense requires real-time visibility across all endpoints, according to Tanium CEO Dan Streetman. He shared how Tanium's "confidence score" framework enables organizations to monitor operational impact on every endpoint when a change is rolled out, helping teams remediate threats at scale.
Anthropic's Jason Clinton Discusses the Benefits and Challenges of AI AgentsAI agents will be crucial in the software development life cycle to eliminate bugs, improving the quality of software, which could significantly reduce security vulnerabilities. Although managing AI agents for identity and access controls will be hard, said Jason Clinton, CISO at Anthropic.
DistributedApps.ai's Ken Huang on Agentic AI Risks and Threat ModelingAs AI agents gain autonomy and access dynamic tools, organizations must adopt new threat modeling approaches like mixture threat modeling, a new method that accounts for AI's unpredictability, said Ken Huang, chief AI officer at DistributedApps.ai.
Menlo Venture's Rama Sekhar on AI Threats and OpportunitiesPublic attention has been focused on the dangers of large language models such as hallucinations or harmful output, but the most pressing security risks are no longer rooted in the models, but in how they are integrated with real-world tools, said Rama Sekhar, partner at Menlo Ventures.
Real-time video deepfakes? Not convincing yet RSAC Spam messages predate the web itself, and generative AI has given it a fluency upgrade, churning out slick, localized scams and letting crooks hit regions and dialects they used to ignore.…
Run by the team at workflow orchestration and AI platform Tines, the Tines library features pre-built workflows shared by security practitioners from across the community - all free to import and deploy through the platform’s Community Edition
A employee at Elon Musk's artificial intelligence company xAI leaked a private key on GitHub that for the past two months could have allowed anyone to query private xAI large language models (LLMs) which appear to have been custom made for working with internal data from Musk's companies, including SpaceX, Tesla and Twitter/X, KrebsOnSecurity has learned.
This year's top cyber challenges include cloud authorization sprawl, ICS cyberattacks and ransomware, a lack of cloud logging, and regulatory constraints keeping defenders from fully utilizing AI's capabilities.
Palo Alto Networks' Nikesh Arora: Browser Security Will Surpass EDR in ImportanceAs enterprises rush to deploy AI across operations, Palo Alto Networks is securing models and agents through its platform approach and recent acquisitions. CEO Nikesh Arora predicts browser security will outpaceEDR as a foundational requirement.
CyberArk's Matt Cohen: AI Agents Add New Category to Exploding Identity LandscapeMatt Cohen, CEO of CyberArk, explains how cloud-native applications have exponentially increased machine identities, with AI agents now creating an entirely new identity type requiring similar authentication and life cycle management approaches.
Will the personal assistant shop for groceries? Or get hijacked by a teen? RSAC If Amazon's Alexa+ works as intended, it could show how an AI assistant helps with everyday tasks like making dinner reservations or arranging an oven repair. Or things could go terribly wrong: it might turn on the oven and turn dinner plans into a house fire.…
Will the personal assistant shop for groceries? Or get hijacked by a teen? RSAC If Amazon's Alexa+ works as intended, it could show how an AI assistant helps with everyday tasks like making dinner reservations or arranging an oven repair. Or things could go terribly wrong: it might turn on the oven and turn dinner plans into a house fire.…
Zscaler's Jay Chaudhry on Visibility and Policy Enforcement to Secure AI UsageOrganizations face significant risks when employees use public AI tools without governance, but security platforms can provide visibility, policy controls and data protection to safeguard sensitive information from unauthorized exposure, said Jay Chaudhry, founder, chairman and CEO, Zscaler.
Anthropic has found its Claude chatbot is being used for automated political messaging, enabling AI-driven influence campaigns