Sharing Information and Intelligence without Disclosing It - Private Search Set (PSS)
summary
Private Search Set (PSS) packages serialized Bloom filters and small metadata so teams can query indicators locally without revealing raw values, add watermark via HMAC key to trace leaks, and tune false‑positive rate by sizing filters or combining bloom + explicit lists. Implementations: dcso Bloom reimplementation (Rust/poy), a Python meta‑lib and a YARA module; deployed examples include a 1 GB ASH lookup bloom, darknet proxy blocking and bank‑account sharing—real builds that work in production.
AI-written summary · about
chapters (13)
- 0:00Private Search Set intro
- 3:12Indicator sharing in CTI
- 6:03Bloom filters for security
- 9:15Ash Lookup use case
- 11:21PSS for Yara and L Project
- 13:27Standardizing PSS format
- 17:03Bloom filter implementation
- 18:32Dark web gateway use case
- 20:42Financial sector with PSS
- 21:46Monitoring keywords in L Project
- 22:28Future plans for PSS
- 24:12Conclusion and open-source
- 25:16Q&A on Bloom filter tuning
related talks
31mPractical Private DB Querying
36mThe Forbidden Index: How We Turned OpenSearch Into a Privacy-Preserving Superhero
19mPrefix Siphoning: Exploiting LSM-Tree Range Filters For Information Disclosure
46mPrivacy Preserving IoT
35mHow to Secure Your Software Supply Chain and Speed-Up DFIR with Hashlookup
40mKeeping data private in real-time pipelines
19mPSSH, or the Primordial Soup of Secure Headers
29mPULSE: Parallel Private Set Union for Large-Scale Entities