What changed in AI security, Apr 7 to Apr 13, 2025
Apr 7 to Apr 13, 2025 (ISO week 2025-W15). Weeks run Monday to Sunday in UTC.
8 records published, -2 on the previous week: 5 vulnerabilities (-4), 0 incidents (no change), 1 research item (+1), 2 news items (+2), 0 policy items (-1).
Critical and high advisories
Vulnerability records rated critical or high, newest first.- Critical
CVE-2025-32375: BentoML insecure deserialization in runner server via crafted POST request
CVE-2025-32375NVD/CVE Database - High
CVE-2025-32018: Cursor agent writes to files outside the opened workspace
CVE-2025-32018NVD/CVE Database - Critical
CVE-2025-3248: Langflow code injection in /api/v1/validate/code endpoint
CVE-2025-3248NVD/CVE Database
Exploitation signals
Vulnerabilities published in the week that are listed in the CISA Known Exploited Vulnerabilities catalog or have an EPSS score of 10% or more.| Advisory | Exploitation | EPSS | Published |
|---|---|---|---|
| CVE-2025-3248: Langflow code injection in /api/v1/validate/code endpoint CVE-2025-3248NVD/CVE Database | Known exploited, ransomware use | 100.0% | |
| CVE-2025-32375: BentoML insecure deserialization in runner server via crafted POST request CVE-2025-32375NVD/CVE Database | Not listed | 52.4% |
Packages that began delegating to a language model
Exposure Registry packages whose first release declaring an LLM SDK, agent framework or MCP dependency was published in the week.| Package | Ecosystem | LLM SDKs | Release | Released |
|---|---|---|---|---|
| openinference-instrumentation | PyPI | OpenAI SDK | 0.1.27 |
Topics that moved
Largest increases over the mean of the 4 previous weeks, for topics with at least 3 records in the week.No topic had at least 3 records in this week and more than its mean over the 4 previous weeks.
Research
Peer-reviewed first, then newest.Policy and regulation
Newest first.No regulatory or policy records were published in this week.
Generated from the AI Sec Watch database at . Every item links to its record.