aisecwatch.com
DashboardVulnerabilitiesNewsResearchArchiveStatsDatasetFor devs
Subscribe
aisecwatch.com

Real-time AI security monitoring. Tracking AI-related vulnerabilities, safety and security incidents, privacy risks, research developments, and policy changes.

Navigation

VulnerabilitiesNewsResearchDigest ArchiveNewsletter ArchiveSubscribeData SourcesStatisticsDatasetAPIIntegrationsWidgetRSS Feed

Maintained by

Truong (Jack) Luu

Information Systems Researcher

Browse All

All tracked items across vulnerabilities, news, research, incidents, and regulatory updates.

to
Export CSV
6364 items

Micron rides memory price spike into earnings with stock up 62%, drubbing its tech peers

infonews
industry
Mar 18, 2026

Micron Technology's stock has surged 62% in 2026 due to a severe shortage of memory chips (computer components that store data temporarily) needed for AI graphics processing units (GPUs, specialized chips that power artificial intelligence). The shortage is driven by massive demand from cloud companies like Amazon and Google building AI data centers, and SK Hynix estimates the memory crunch will continue for another four to five years, pushing prices higher across the industry.

CNBC Technology

We asked experts about the most responsible ways to use AI tools – here’s what they said

infonews
safety
Mar 18, 2026

The article discusses expert advice on responsible AI tool use, emphasizing that people should use AI as a brainstorming partner and for organizing information, but should not let it replace their own decision-making. A 2025 survey shows that one-third of US adults use ChatGPT, with particularly high adoption among people under 30.

Can you prove the person on the other side is real?

infonews
securitysafety

China’s ‘AI tigers’ see shares surge after Nvidia CEO touts OpenClaw as ‘next ChatGPT’

infonews
industry
Mar 18, 2026

Chinese AI companies saw significant stock gains after Nvidia CEO Jensen Huang praised OpenClaw, an open-source AI agent (a program that can perform tasks independently), as "the next ChatGPT." Companies like MiniMax and Zhipu, which are among China's leading AI developers building large language models (AI systems trained on huge amounts of text to understand and generate language), have integrated OpenClaw into their products and are launching their own versions based on it.

CISOs rethink their data protection strategies

infonews
securitypolicy

Meta's Manus launches desktop app to bring its AI agent onto personal devices amid OpenClaw craze

infonews
industrysafety

OWASP GenAI Security Project Expands AI Security Frameworks Ahead of RSA 2026, Celebrates Continued Sponsor Support

inforesearchIndustry
security

Survey on Learning-based Dynamic Fault Localization: From Traditional Machine Learning to Large Language Models

inforesearchPeer-Reviewed
research

Die besten Hacker-Filme

infonews
security
Mar 18, 2026

This is a curated list of hacker-themed films arranged chronologically, from War Games (1983) to Live Free or Die Hard (2007), intended for security professionals who enjoy cinema. The article provides plot summaries, genres, and review scores from multiple sources for each film, with a note that the list may cause procrastination.

CVE-2026-20963: Microsoft SharePoint Deserialization of Untrusted Data Vulnerability

infovulnerability
security
Mar 17, 2026
CVE-2026-20963

Microsoft SharePoint has a deserialization of untrusted data vulnerability (a flaw where the software unsafely processes data from untrusted sources, allowing attackers to inject malicious code). An unauthorized attacker can exploit this over a network to execute code on affected systems. This vulnerability is currently being actively exploited by real attackers.

CVE-2025-66376: Synacor Zimbra Collaboration Suite (ZCS) Cross-Site Scripting Vulnerability

infovulnerability
security
Mar 17, 2026
CVE-2025-66376🔥 Actively Exploited

Nvidia CEO Jensen Huang says OpenClaw is 'definitely the next ChatGPT'

infonews
industry
Mar 17, 2026

Nvidia CEO Jensen Huang highlighted OpenClaw, an open-source autonomous AI agent platform (a system that can complete tasks and make decisions with minimal human input, unlike traditional chatbots), calling it "the next ChatGPT" and a major breakthrough in AI interaction. Nvidia launched NemoClaw, an enterprise version of OpenClaw that adds security, scalability, and oversight tools to make these autonomous agents safe for real-world business use, addressing concerns about security, privacy, and control as these systems gain the ability to act independently.

The Pentagon is planning for AI companies to train on classified data, defense official says

infonews
policysecurity

OpenAI preps for IPO by end of year, tells employees ChatGPT must be 'productivity tool'

infonews
industry
Mar 17, 2026

OpenAI is preparing for an initial public offering (IPO, where a private company sells shares to the public) potentially by the end of 2024, with leadership telling employees that ChatGPT must focus on being a productivity tool for businesses. The company is shifting strategy to convert its 900 million weekly users into enterprise customers and has scaled back its infrastructure spending targets from $1.4 trillion to $600 billion by 2030 to present a more realistic financial picture to investors.

GHSA-2cpp-j2fc-qhp7: AWS API MCP File Access Restriction Bypass

mediumvulnerability
security
Mar 17, 2026
CVE-2026-4270

The AWS API MCP Server (a tool that lets AI assistants interact with AWS services) has a vulnerability in versions 0.2.14 through 1.3.8 where attackers can bypass file access restrictions and read files they shouldn't be able to access, even when the server is configured to block file operations or limit them to a specific directory.

GHSA-9x67-f2v7-63rw: AVideo vulnerable to unauthenticated SSRF via HTTP redirect bypass in LiveLinks proxy

highvulnerability
security
Mar 17, 2026
CVE-2026-33039

AVideo's LiveLinks proxy endpoint validates URLs to block requests to internal networks, but only checks the initial URL. When a URL redirects (sends back a `Location` header pointing elsewhere), the code follows the redirect without re-validating the new target, letting attackers reach internal services like cloud metadata or private networks. The endpoint is also completely unauthenticated, so anyone can access it.

GHSA-vwmf-pq79-vjvx: Unauthenticated Remote Code Execution in Langflow via Public Flow Build Endpoint

criticalvulnerability
security
Mar 17, 2026
CVE-2026-33017

Langflow has an unauthenticated remote code execution vulnerability in its public flow build endpoint. The endpoint is designed to be public but incorrectly accepts attacker-supplied flow data containing arbitrary Python code, which gets executed without sandboxing when the flow is built. An attacker only needs to know a public flow's ID and can exploit this to run any code on the server.

GHSA-2f9h-23f7-8gcx: AVideo affected by unauthenticated application takeover via exposed web installer on uninitialized deployments

highvulnerability
security
Mar 17, 2026
CVE-2026-33038

AVideo's web installer endpoint (`install/checkConfiguration.php`) allows unauthenticated attackers to fully set up the application on fresh deployments by sending POST requests with attacker-controlled database credentials, admin passwords, and configuration values. Since the only protection is checking if a configuration file exists, attackers can take over uninitialized instances by pointing them to an attacker-controlled database and creating admin accounts with attacker-chosen passwords.

GPT-5.4 mini and GPT-5.4 nano, which can describe 76,000 photos for $52

infonews
industry
Mar 17, 2026

OpenAI released two new smaller AI models, GPT-5.4 mini and GPT-5.4 nano, that are cheaper and faster than previous versions. GPT-5.4 nano is particularly affordable at $0.20 per million input tokens, making it economical for tasks like image description, where describing 76,000 photos would cost around $52.

Nvidia NemoClaw promises to run OpenClaw agents securely

infonews
securityindustry
Previous159 / 319Next
The Guardian Technology
Mar 18, 2026

Synthetic identity fraud, where criminals create fake people using AI-generated documents and deepfakes (realistic fake videos or audio), is becoming a major threat in estate and identity verification work. Traditional security checks that look at device fingerprints or typing patterns are no longer reliable because AI can now imitate these signals. The text explains that the real challenge by 2026 will be distinguishing legitimate people from manufactured personas, especially in high-stakes situations involving inheritance and family claims.

Fix: The source suggests moving from asking "Who is this?" to a more forensic approach: "How did this identity—and its digital footprint—come to exist?" This shift means prioritizing provenance (where the identity originated), issuer verification (confirming documents are real), and cross-channel consistency (checking if the person's presence makes sense across multiple systems) over accepting surface-level plausibility. However, the text does not provide specific technical implementations or detailed steps for executing this approach.

CSO Online
CNBC Technology
Mar 18, 2026

CISOs (Chief Information Security Officers, the top security leaders at companies) are updating their data protection strategies because employees are rapidly sharing company data with AI tools, including public models like ChatGPT, creating new security risks. A CISO at a law firm added a new protection layer that classifies data based on whether it can be safely used with AI and invested in new monitoring tools, while also regularly evaluating new technologies to ensure controls keep pace with AI innovations.

Fix: The source describes one organization's approach: add a protection layer that classifies and tags data based on whether it could be used with AI and in what circumstances, invest in new tools to support that layer, monitor the vendor landscape for emerging capabilities, and evaluate new technologies being deployed to determine whether new controls are needed for them. However, no specific technical solutions, patches, or vendor recommendations are explicitly named in the source text.

CSO Online
Mar 18, 2026

Meta-owned Manus launched a desktop application with a feature called 'My Computer' that allows its AI agent (a program that can complete complex, multi-step tasks automatically) to access and control files, tools, and applications directly on a user's computer, rather than only working in the cloud. This move competes with OpenClaw, a free, open-source AI agent that similarly runs on local devices. Experts have raised security and privacy concerns about giving AI agents local device access, but Manus addressed this by requiring explicit user approval before the agent executes tasks.

Fix: Manus's mitigation for security and privacy risks includes a control mechanism requiring explicit user approval before task execution. According to Manus, users can choose "Allow Once" for individual review of each action or "Always Allow" for trusted, recurring actions, keeping users "firmly in control."

CNBC Technology
policy
Mar 18, 2026

The OWASP GenAI Security Project, an open-source community focused on AI security, announced expansion of its resources and frameworks with over 25,000 members contributing practical guidance and tools. The project is being highlighted at the RSA 2026 conference, indicating growing industry adoption of AI security best practices.

OWASP GenAI Security
Mar 18, 2026

This survey examines methods for automatically finding bugs in software code by using machine learning and AI models, tracing the evolution from traditional machine learning techniques to modern large language models (LLMs, which are AI systems trained on vast amounts of text data). The research covers how these AI-based approaches learn patterns to pinpoint where faults occur in code, making debugging faster and more efficient than manual inspection.

ACM Digital Library (TOPS, DTRAP, CSUR)
CSO Online

Fix: Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable. Due date: 2026-03-21.

CISA Known Exploited Vulnerabilities

Zimbra Collaboration Suite (ZCS) has a cross-site scripting vulnerability (XSS, a type of attack where malicious code runs in a user's browser) in its Classic UI that allows attackers to exploit CSS @import directives (special commands that load external stylesheets) in email HTML. This vulnerability is currently being actively exploited by attackers in real-world attacks.

Fix: Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable. The due date for remediation is 2026-04-01.

CISA Known Exploited Vulnerabilities

Fix: Nvidia addressed risks with NemoClaw by building "guardrails, including privacy protections, oversight tools, and enterprise-grade security to ensure these agents can be deployed safely at scale."

CNBC Technology
Mar 17, 2026

The Pentagon is planning to let AI companies train their models on classified military data in secure facilities, which would allow the AI to learn from and embed sensitive intelligence like surveillance reports. While this could make AI systems more accurate for military tasks, experts warn it creates risks: classified information that the AI learns could accidentally be shared with people or military departments that shouldn't have access to it, potentially endangering operatives or exposing secrets.

MIT Technology Review
CNBC Technology

Fix: Upgrade to version 1.3.9 or later.

GitHub Advisory Database
GitHub Advisory Database
GitHub Advisory Database
GitHub Advisory Database
Simon Willison's Weblog
Mar 17, 2026

OpenClaw, a framework for running AI agents (autonomous programs that can take actions) locally on devices rather than in the cloud, has faced security concerns since its rapid rise in early 2026. Nvidia announced NemoClaw, which addresses these vulnerabilities by using OpenShell, a security layer that includes kernel-level sandboxing (isolating programs from the core system) and a privacy router that monitors and blocks unauthorized data transfers by OpenClaw.

Fix: NemoClaw's OpenShell runtime isolates OpenClaw using kernel-level sandboxing and a 'privacy router' that monitors OpenClaw's behavior and communication with other systems, stepping in to block actions if it detects OpenClaw sending sensitive data somewhere it shouldn't. OpenShell is fully open source.

CSO Online