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.
Over 30 security vulnerabilities have been disclosed in various artificial intelligence (AI)-powered Integrated Development Environments (IDEs) that combine prompt injection primitives with legitimate features to achieve data exfiltration and remote code execution
A campaign has been observed targeting Palo Alto GlobalProtect portals with login attempts and launching scanning activity against SonicWall SonicOS API endpoints. [...]
Enterprises Are Reimagining Org Roles, Risk Management and Skillsets in the AI RaceOrganizations are beginning to reimagine how leadership roles should be structured, aligned and empowered as they grapple with regulatory pressures, the unpredictable nature of AI systems, and the need for operational resilience in an increasingly uncertain business climate.
Closed AI Loops Are Concentrating Power - and Creating Room for StartupsMicrosoft, Nvidia and Anthropic just formed the latest closed-loop artificial intelligence partnership, tying cloud, hardware and models into a single circuit. While it signals consolidation at the top, founders say it's also creating a surprising tailwind for domain-focused AI startups.
Also: More HIPAA Challenges, the Growing AI Gap for Small- to Medium-Sized FirmsIn this week's panel, four ISMG editors discussed the latest shifts in ransomware tactics, a major development in the Texas challenge to the HIPAA Privacy Rule related to reproductive rights, and how SMBs navigating AI are facing very different challenges than large enterprises.
Automated software keeps getting better at pilfering cryptocurrency Anthropic could have scored an easy $4.6 million by using its Claude AI models to find and exploit vulnerabilities in blockchain smart contracts.…
Plan Aims to Modernize Workflow, Expand AI Use Across Agencies, Improve CyberThe U.S. Department of Health and Human Services on Thursday unveiled "version 1" of a strategic plan to implement artificial intelligence as a "practical layer" across the department and its agencies aimed at helping to break down silos, improve collaboration and increase efficiencies.
Transurban head of cyber defense Muhammad Ali Paracha shares how his team is automating the triaging and scoring of security threats as part of the Black Hat Middle East conference.
Global cybersecurity agencies published guidance regarding AI deployments in operational technology, a backbone of critical infrastructure.
AI in OT May Trigger Cascading Infrastructure FailuresThe U.S. cyber defense agency warned that machine learning and large language model deployments can introduce new attack surfaces across critical infrastructure sectors in a document setting out principles for safely integrating AI into operational technology.
And then they asked an AI to help cover their tracks Vetting staff who handle sensitive government systems is wise, and so is cutting off their access the moment they're fired. Prosecutors say a federal contractor learned this the hard way when twin brothers previously convicted of hacking-related offenses allegedly used lingering access to delete nearly 100 government databases, including systems tied to Homeland Security and other agencies, within minutes of being terminated.…
Also: Anthropic Warns of Autonomous AI Exploits on BlockchainThis week, authorities shutter Cryptomixer, Anthropic warns about autonomous AI exploits, U.K. plans ban on crypto political donations, Do Kwon seeks leniency, Lazarus Group suspected in Upbit theft, Balancer's post-exploit plans and Yearn recovers some hacked amount.
Cybersecurity agencies have issued guidance for securely integrating AI into OT systems
The deal, believed to be valued at $1 billion, will bring non-human identity access control of agents and machines to ServiceNow’s offerings including its new AI Control Tower.
UK’s NCSC and Canada’s CCCS release a joint report on content provenance, urging organizations to strengthen digital trust and combat AI-driven misinformation
As 2025 draws to a close, security professionals face a sobering realization: the traditional playbook for web security has become dangerously obsolete. AI-powered attacks, evolving injection techniques, and supply chain compromises affecting hundreds of thousands of websites forced a fundamental rethink of defensive strategies
China Still Relies on US Technology, Experts tell SenateWashington spent years constructing export barriers around America's most sensitive artificial intelligence technology. Witnesses told the U.S. Senate Foreign Relations Committee that China is finding ways to move around them. Where one pathway closes, Beijing opens another.
Attackers Could Hijack Developer Machines via Tampered Config FilesOpenAI patched a command injection flaw in its Codex CLI tool that let attackers run arbitrary commands on developer machines by hiding malicious configuration files inside code repositories. Hackers could turn ordinary repository files into execution vectors.
Organizations are adopting agentic artificial intelligence as the next phase of AI. Kim Basile, CIO of Kyndryl, explains how organizations can prepare teams to work with agentic AI, emphasizing culture, training and governance as the crucial drivers of AI readiness and adoption.
Remember when phishing emails were easy to spot? Bad grammar, weird formatting, and requests from a "Prince" in a distant country? Those days are over