AI security

AI security refers to the practices, technologies, and policies designed to protect artificial intelligence systems from threats, vulnerabilities, and malicious attacks. It involves securing the data, models, and infrastructure used in AI systems to ensure their integrity, confidentiality, and availability. AI security also includes defending against adversarial attacks, preventing data poisoning, safeguarding against model theft or misuse, and ensuring that AI systems operate safely and reliably under potential adversarial conditions.
Recent contents
Recent content 14 News story 12 Link directory item 2
  1. Google AVDH Brings Agentic AI to Enterprise Source Code Review

    Google AVDH Brings Agentic AI to Enterprise Source Code Review

    Mandiant turns AI code review into a gated workflow Google Cloud Threat Intelligence has described AVDH, an internal Mandiant system that uses agentic AI to review source code during proactive assessments, red team work and incident response. The post frames the harness as a defensive answer to...
  2. GLM-5.3 brings open-weight AI closer to cyber exploitation

    GLM-5.3 brings open-weight AI closer to cyber exploitation

    Z.ai Is Preparing to Release an AI Model With Serious Cyber Capabilities Chinese AI company Z.ai has unveiled GLM-5.3, a new coding model whose most consequential upgrade may be its ability to find and investigate software vulnerabilities. Z.ai describes the model's cybersecurity behavior as an...
  3. Agentic security pushes Google red teams toward AI drills

    Agentic security pushes Google red teams toward AI drills

    Google Reframes Red Teaming for Autonomous Threats Google says the red-team mission is not changing, but the operating model has to adapt as AI agents alter attacker speed, scale and cost. In a Security blog post, the company argues that defenders should test whether monitoring, response and...
  4. OpenAI Daybreak Models Reach Amazon Bedrock for Cyber Defense

    OpenAI Daybreak Models Reach Amazon Bedrock for Cyber Defense

    AWS Adds Governed OpenAI Cyber Models to Bedrock AWS says OpenAI’s Daybreak Red and Daybreak Blue are now available on Amazon Bedrock for eligible customers, bringing specialized cyber defense models into a managed cloud AI service. The launch is aimed at security teams that need AI assistance...
  5. How AgentCore temporal policies govern AI agent workflows

    How AgentCore temporal policies govern AI agent workflows

    Stateful policy controls move closer to agent gateways AWS has detailed temporal policies for Amazon Bedrock AgentCore, a governance feature aimed at AI agents that choose tools and arguments at runtime. The policies evaluate an agent's current request against earlier events in the same session...
  6. Apple DLR-Lock research targets open-weight model tuning

    Apple DLR-Lock research targets open-weight model tuning

    Apple researchers test a technical lock for fine-tuning LLM weights Apple Machine Learning Research has published work on DLR-Lock, a method intended to make open-weight language models harder to adapt after release. The paper frames the issue as a tradeoff: shared weights support adoption...
  7. AI token jacking turns stolen API keys into gray market access

    AI token jacking turns stolen API keys into gray market access

    Stolen AI API Keys Become a Billing and Abuse Risk Unit 42 says cybercriminals are stealing developer API keys for AI platforms and using them to consume or resell model access. The security team describes the activity as token jacking, an AI-focused version of stealing access to paid computing...
  8. AI vulnerability discovery finds 14,090 open-source flaws

    AI vulnerability discovery finds 14,090 open-source flaws

    Autonomous AI vulnerability research moves from theory to scale Unit 42 says an autonomous research system called NOVA confirmed 14,090 vulnerabilities across 3,915 open-source projects in two months. The company describes the work as evidence that frontier AI can expand vulnerability discovery...
  9. Protect AI

    Protect AI

    Overview Protect AI is an AI security platform focused on discovering, assessing and reducing risks across machine-learning systems and supply chains. Features, limits and availability should be confirmed on the official site. Best for Security teams assessing AI and machine-learning...
  10. Zenity

    Zenity

    Overview Zenity is an enterprise security and governance platform for discovering, assessing and protecting AI agents across SaaS, cloud and end-user environments. It combines observability, exposure management, identity controls and response, but deployment requires mature security processes...
  11. OpenAI Fixes ChatGPT’s Long Em Dash Problem

    OpenAI Fixes ChatGPT’s Long Em Dash Problem

    OpenAI Removes Forced Em Dash: ChatGPT Gains More Natural Writing Control OpenAI quietly fixed a long-standing punctuation problem that frustrated millions of ChatGPT users - the model finally stops forcing the long em dash even when explicitly asked not to. A subtle symbol that became an AI...
  12. Anthropic Stops First AI-Driven Cyber Espionage Operation GTG-1002

    Anthropic Stops First AI-Driven Cyber Espionage Operation GTG-1002

    Anthropic Disrupts First AI-Driven Cyber Espionage Campaign GTG-1002 Anthropic’s threat intelligence team confirmed that it has stopped the first known case of a state-backed cyber espionage campaign executed predominantly by an artificial intelligence system. The operation, labelled GTG-1002...
  13. Longer AI Reasoning Makes Models Easier to Jailbreak, Study Finds

    Longer AI Reasoning Makes Models Easier to Jailbreak, Study Finds

    Longer AI Reasoning Makes Models More Vulnerable to Jailbreaks, Researchers Warn A new joint study by Anthropic, Stanford University, and the University of Oxford challenges one of the core assumptions in modern AI safety: that extending a model’s reasoning time makes it harder to exploit...
  14. Safeway Installs Anti-Theft Exit Gates: Shoppers Can’t Leave Without Paying

    Safeway Installs Anti-Theft Exit Gates: Shoppers Can’t Leave Without Paying

    Safeway Installs Anti-Theft Exit Gates: Shoppers Can’t Leave Without Paying In a move that feels straight out of a dystopian future, a Safeway supermarket in San Francisco has introduced a security system that physically prevents customers from leaving the store unless they’ve made a purchase...
  • This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Accept Learn more…
Back
Top