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
6268 items

GHSA-38h3-2333-qx47: OpenTelemetry .NET has potential memory exhaustion via unbounded pooled-list sizing in Jaeger exporter conversion path

mediumvulnerability
security
Apr 17, 2026
CVE-2026-41078

OpenTelemetry.Exporter.Jaeger has a memory exhaustion vulnerability where internal pooled lists (reusable memory structures) can grow too large based on big payloads and stay oversized for future use, potentially causing denial of service (making a system unavailable). However, the developers have no plans to fix this because the Jaeger exporter was deprecated in 2023.

Fix: Prefer maintained exporters (for example OpenTelemetry Protocol format (OTLP)) instead of the Jaeger exporter.

GitHub Advisory Database

GHSA-v38x-c887-992f: Flowise: Airtable_Agent Code Injection Remote Code Execution Vulnerability

criticalvulnerability
security
Apr 17, 2026

Flowise versions up to 3.0.13 have a remote code execution vulnerability in the Airtable Agent node where user input is sent to an LLM (large language model, an AI that generates text) to generate Python code, which is then executed without proper sandboxing. An attacker can craft malicious prompts that trick the LLM into generating code containing dangerous commands (like imports or system operations) that bypass the validation checks, allowing them to run arbitrary code on the server without needing to log in.

White House and Anthropic hold 'productive' meeting amid fears over Mythos model

infonews
policyindustry

BioGuard: Malicious sample free defense method for biometric classifiers against model extraction attacks

inforesearchPeer-Reviewed
security

OpenAI loses multiple executives in latest leadership shakeup

infonews
industry
Apr 17, 2026

OpenAI experienced multiple executive departures, including the leaders of its video generation product (Sora) and its scientific research division. The company is reorganizing its science team to work more closely with product and infrastructure groups, while also dealing with medical leaves and transitions among other senior leaders.

AI chipmaker Cerebras files to go public after scrapping IPO plans last year

infonews
industry
Apr 17, 2026

Cerebras, a company that makes specialized chips for running AI models, filed to go public on Nasdaq after previously canceling IPO plans in 2024. The company reported strong financial growth in 2025 with $510 million in revenue (up 76% from 2024) and has major deals with OpenAI (worth over $20 billion for computing power through 2028) and Amazon, positioning itself as an alternative to Nvidia's GPUs (graphics processing units, specialized processors commonly used for AI tasks) by claiming faster speeds and lower costs.

Breaking Opus 4.7 with ChatGPT (Hacking Claude's Memory)

infonews
securitysafety

GHSA-8gmg-3w2q-65f4: OpenTelemetry eBPF Instrumentation: Privileged Java agent injection allows arbitrary host file overwrite via untrusted TMPDIR

highvulnerability
security
Apr 17, 2026

OpenTelemetry eBPF Instrumentation (OBI) has a vulnerability where a local attacker controlling a Java process can overwrite arbitrary host files when Java injection is enabled and OBI runs with elevated privileges (special system permissions). The flaw occurs because the injector trusts an environment variable called TMPDIR from the target process without proper validation, and uses unsafe file creation methods that allow symlink attacks (where an attacker creates a link pointing to a different file to trick the system into overwriting it).

GHSA-5cwg-9f6j-9jvx: Claude Code: Insecure System-Wide Configuration Loading Enables Local Privilege Escalation on Windows

mediumvulnerability
security
Apr 17, 2026
CVE-2026-35603

Claude Code on Windows had a security flaw where it loaded configuration files from a shared system directory without checking who owned that directory or had permission to change it. Since regular users could write to this directory by default, an attacker could create a malicious configuration file that would run with elevated privileges when another user launched Claude Code, allowing a local privilege escalation (unauthorized access to higher-level permissions).

GHSA-66r7-m7xm-v49h: OpenClaw: QQBot media tags could read arbitrary local files through reply text

highvulnerability
security
Apr 17, 2026

QQBot media tags in the openclaw package could read arbitrary local files through reply text by referencing host-local paths outside the intended media storage boundary, allowing attackers to disclose local files through outbound media handling. This vulnerability affected openclaw versions before 2026.4.10.

CVE-2026-40352: FastGPT is an AI Agent building platform. In versions prior to 4.14.9.5, the password change endpoint is vulnerable to N

highvulnerability
security
Apr 17, 2026
CVE-2026-40352

FastGPT, an AI Agent building platform, has a vulnerability in its password change feature in versions before 4.14.9.5 where attackers can use NoSQL injection (inserting MongoDB operators into input fields to manipulate database queries) to bypass password verification and take over accounts without knowing the current password.

CVE-2026-40351: FastGPT is an AI Agent building platform. In versions prior to 4.14.9.5, the password-based login endpoint uses TypeScri

criticalvulnerability
security
Apr 17, 2026
CVE-2026-40351

FastGPT, an AI Agent building platform, has a NoSQL injection vulnerability (a type of attack where an attacker tricks the database query by inserting special commands) in its login system before version 4.14.9.5. The vulnerability allows unauthenticated attackers to bypass password checks and log in as any user, including administrators, by sending database operators instead of a real password.

GHSA-vfp4-8x56-j7c5: OpenClaw: Exec environment denylist missed high-risk interpreter startup variables

highvulnerability
security
Apr 17, 2026

OpenClaw missed blocking dangerous environment variables (like VIMINIT, EXINIT, LUA_INIT, and HOSTALIASES) that could be set by users to change how programs start up or behave on the network. This security gap affected OpenClaw versions before 2026.4.10.

GHSA-5fw2-mwhh-9947: Flowise: Unauthenticated TTS endpoint accepts arbitrary credential IDs — enables API credit abuse via stored credentials

highvulnerability
security
Apr 17, 2026

Flowise has a text-to-speech endpoint that doesn't require authentication but accepts a credential ID (an identifier for stored API keys like OpenAI or ElevenLabs) directly from user input. An attacker can use this to access someone else's stored API credentials and generate speech using the victim's API account, burning their API credits without permission.

GHSA-w47f-j8rh-wx87: Flowise: Public chatflow endpoints return unsanitized flowData including plaintext API keys, passwords, and credential IDs

highvulnerability
security
Apr 17, 2026

Flowise version 3.0.13 has a security flaw where public chatflow endpoints return unsanitized data (raw information without filtering) that includes plaintext API keys, passwords, and credential IDs (unique references to stored login credentials). This happens because the code returns the complete chatflow object without removing sensitive fields, potentially exposing users' third-party account credentials and internal system architecture.

OpenAI’s former Sora boss is leaving

infonews
industry
Apr 17, 2026

OpenAI abandoned its Sora video generation tool and Bill Peebles, the leader of the Sora team, is leaving the company. OpenAI is refocusing its priorities away from what it calls "side quests" to concentrate on coding and enterprise products instead.

Should you stare into Sam Altman’s orb before your next date?

infonews
security
Apr 17, 2026

Tinder is partnering with World, a company co-founded by OpenAI CEO Sam Altman, to let users verify their identity using facial scanning orbs (physical devices that take pictures of faces and eyes to confirm someone is a real person, not a bot or AI agent). Users who complete this verification in select markets like Japan and the United States will receive five free boosts in the app.

Anthropic’s new cybersecurity model could get it back in the government’s good graces

infonews
industrypolicy

Perspective: AI demand is inflated, and only Anthropic is being realistic

infonews
industry
Apr 17, 2026

AI companies may be overestimating demand by measuring success through token consumption (the basic units of AI usage, like words and characters), rather than actual business value or return on investment. Anthropic is adjusting its pricing model away from flat monthly fees toward per-token billing and has discontinued third-party tools that were consuming excessive tokens without generating meaningful results, positioning itself better if AI demand projections prove inflated.

White House Chief of Staff to Meet With Anthropic CEO Over Its New AI Technology

infonews
policy
Apr 17, 2026

The White House is planning a meeting between its Chief of Staff and Anthropic's CEO to discuss Anthropic's new AI technology and concerns about the security of software built with advanced AI models. This reflects ongoing government engagement with major AI labs about how their systems work and potential risks.

Previous108 / 314Next
GitHub Advisory Database
Apr 17, 2026

Anthropic, an AI company, met with White House officials after releasing Claude Mythos, an AI tool that can find bugs in old code and autonomously exploit them for security testing. The meeting signals potential collaboration between the government and Anthropic despite previous tensions, as officials discussed balancing innovation with safety concerns around this powerful technology.

BBC Technology
research
Apr 17, 2026

Researchers have developed BioGuard, a defense method that protects biometric classifiers (AI systems that identify people using fingerprints, faces, or iris scans) against model extraction attacks (where attackers try to steal or copy the AI model by repeatedly querying it). The method works without needing malicious sample data to train it, making it practical for real-world deployment.

Elsevier Security Journals
CNBC Technology
CNBC Technology
Apr 17, 2026

A researcher discovered that Claude Opus 4.7 can be tricked using an adversarial image (a specially crafted image designed to fool AI systems) generated by ChatGPT to misuse the memory tool and store false information for future conversations. While Claude Opus 4.6+ is harder to attack than earlier versions because it reasons through requests before acting, it remains vulnerable to this type of indirect prompt injection (embedding hidden malicious instructions in images rather than text).

Embrace The Red

Fix: Upgrade to https://github.com/open-telemetry/opentelemetry-ebpf-instrumentation/releases/tag/v0.8.0.

GitHub Advisory Database

Fix: Users on standard Claude Code auto-update have already received this fix. Users performing manual updates are advised to update to the latest version.

GitHub Advisory Database

Fix: Upgrade to openclaw version 2026.4.10 or newer. The latest npm release, 2026.4.14, already includes the fix. The fix enforces the media storage boundary for all outbound QQBot local file paths, which was implemented in PR #63271.

GitHub Advisory Database

Fix: Update FastGPT to version 4.14.9.5 or later, where this issue has been fixed.

NVD/CVE Database

Fix: This issue has been fixed in version 4.14.9.5. Users should upgrade to this version or later.

NVD/CVE Database

Fix: Users should upgrade to openclaw version 2026.4.10 or newer. The latest npm release, openclaw@2026.4.14, already includes the fix, which expands the denylist (a list of blocked items) in the execution environment security policy to cover these high-risk environment variables.

GitHub Advisory Database

Fix: Remove the TTS endpoint from the whitelist (the list of endpoints that don't need login), or add a check to ensure the credential ID matches the chatflow's TTS configuration. The source suggests: 'if (!chatflowId) { return res.status(401).json({ message: "Authentication required" }) }' — meaning if no chatflow ID is provided, the endpoint should reject the request with an authentication error.

GitHub Advisory Database

Fix: According to the source, apply sanitization to both public endpoints by calling `sanitizeFlowDataForPublicEndpoint(chatflow)` before returning the response, and ensure the sanitization function removes all `credential`, `password`, `apiKey`, and `secretKey` fields from the flowData. The source notes this sanitization function exists only in unreleased HEAD code, not in released v3.0.13.

GitHub Advisory Database
The Verge (AI)
The Verge (AI)
Apr 17, 2026

Anthropic, an AI company, faced criticism from the Trump administration over concerns about national security and refused to allow its technology to be used for domestic mass surveillance or fully autonomous weapons without human control. The company is now working to improve its relationship with the government by developing Claude Mythos Preview, a new AI model designed specifically for cybersecurity tasks.

The Verge (AI)

Fix: Anthropic's mitigation strategies mentioned in the source include: (1) moving from flat-rate enterprise pricing to per-token billing so revenue reflects actual usage; (2) cutting off third-party agentic tools (like OpenClaw) that were consuming large volumes of tokens unsustainably; and (3) planning infrastructure investment carefully by accounting for a 'cone of uncertainty' (acknowledging that data centers take 1-2 years to build, so companies must estimate future demand carefully rather than over-committing to infrastructure based on inflated projections).

CNBC Technology
SecurityWeek