What changed in AI security, Aug 25 to Aug 31, 2025
Aug 25 to Aug 31, 2025 (ISO week 2025-W35). Weeks run Monday to Sunday in UTC.
13 records published, +3 on the previous week: 2 vulnerabilities (-2), 0 incidents (no change), 3 research items (+3), 8 news items (+2), 0 policy items (no change).
Critical and high advisories
Vulnerability records rated critical or high, newest first.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.No vulnerability published in this week is listed as exploited or has an EPSS score of 10% or more.
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.No tracked package published its first release with an LLM SDK, agent framework or MCP dependency in this week.
Topics that moved
Largest increases over the mean of the 4 previous weeks, for topics with at least 3 records in the week.| Topic | Records | Weekly mean, previous 4 | Difference |
|---|---|---|---|
| Model Context Protocol | 4 | 0.8 | +3.3 |
Research
Peer-reviewed first, then newest.Online Safety Analysis for LLMs: A Benchmark, an Assessment, and a Path Forward
Peer-reviewedIEEE Xplore (Security & AI Journals)Certified Local Transferability for Evaluating Adversarial Attacks
Peer-reviewedIEEE Xplore (Security & AI Journals)Steganography in Large Language Models
Peer-reviewedIEEE Xplore (Security & AI Journals)
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.