Posts tagged: Compliance
Page 5View all articles
-
DeepSeek and the FY2026 NDAA: What DoD Contractors Need to Know
Section 1532 of the FY2026 NDAA bars DeepSeek and High Flyer AI on DoD contracts. Learn what it covers, what it does not, and the replacement path.
-
Using GPT-OSS 20B and 120B for Regulated Industries
GPT-OSS 20B and 120B run on your own hardware under Apache 2.0. Petronella Technology Group, Inc. deploys open-weight models inside compliance boundaries.
-
Why Gemma Is the Right Local AI Model for CMMC Compliance
Run Google's Gemma 4 locally under Apache 2.0 inside your CMMC boundary. Petronella Technology Group, Inc. scopes, sizes, and deploys private AI.
-
Talorys: A Self-Hosted AI Agent on the Cloudflare Free Tier
Talorys is a self-hosted personal AI agent that runs on the Cloudflare free tier. See what its design means for data residency, governance and audits.
-
Agent Corruption Lessons for Designing Safer AI Controls
Memory effects, tool interactions and bad reward signals drive agent corruption, and this guide maps control points and tests that keep AI agents safe.
-
AI Customer Service Handoffs That Get Blocked in Court
AI customer service handoffs become legal evidence in disputes. See the records regulators ask for and the design steps that keep handoffs defensible.
-
Show HN: Jevman - AI decision models play Pac-Man
Jevman benchmark: AI decision models play Pac-Man, showing real-time inference that raises provenance and auditability questions for regulated teams.
-
How a Private AI Voice Agent Cuts Compliance Costs
A private AI voice agent keeps CUI and PHI on hardware you control instead of public cloud, and this guide covers architecture and audit criteria.
-
Federal Contractor Rules for Sensitive Data Near the Finish Line
New federal rules will tighten how contractors protect, monitor and report sensitive data. See a phased path from gap assessment to a formal audit.
-
Consent-First AI Call Scoring That Keeps Trust Intact
AI call scoring built consent first: how to define consent in operational terms, separate recording from AI processing and handle hard edge cases.
-
PoeLLM malware infects exposed AI servers in cryptomining attacks
PoeLLM malware turns exposed AI servers into cryptomining rigs and network scanners. See what regulated teams must monitor and harden to stop it.
-
Oracle Health Data Breach Tally Climbs to Nearly 20 Million
The Oracle Health data breach exposed medical identifiers stored on a vendor cloud; see what it means for third party risk and HIPAA compliance.
