Aligning AI Adoption Maturity with AI Security Using CMMI and Russian Regulatory Requirements
Organizations today operate simultaneously in two distinct dimensions of artificial intelligence: the depth of AI integration into business processes and the maturity of security controls protecting those systems. These two maturity levels rarely align, and the resulting gap is responsible for most security incidents and regulatory findings rather than the mere fact of using AI.
To make the misalignment visible, both axes are mapped onto the five-level CMMI scale. Level L1 represents initial, unpredictable processes; L2 covers managed processes at the project level; L3 introduces organization-wide defined standards; L4 adds quantitative management through metrics; and L5 focuses on continuous optimization.
Under this unified scale, typical adoption follows the Gartner progression from awareness to enterprise transformation, while security maturity is assessed using OWASP AIMA across eight domains including Governance, Data Management, and Operations. The mapping reveals that many organizations reach L3 in adoption while remaining at L1 or L2 in security, leaving them exposed to Shadow AI and violations of mandatory requirements.
FSTEC Order No. 117, effective from March 2026, replaces earlier rules and introduces explicit obligations in paragraphs 60 and 61. These require the use of trusted AI technologies, prohibition on transferring restricted data to model developers, statistical criteria for response accuracy, and controls on query formats. The obligations activate when systems move from pilot to production use, exactly at the L2-to-L3 transition.
Additional Russian anchors include GOST R 56939-2024 for secure development, GOST R 59276-2020 for trust mechanisms, and the new P NST 1046-2026 standard for AI in critical information infrastructure. International references such as Google SAIF, NIST AI RMF, and MITRE ATLAS supply supporting risk maps and attack catalogs that fit within the same CMMI framework.
The practical states organizations pass through include denial of AI use, widespread uncontrolled Shadow AI, LLM controls that leave agentic systems exposed, and finally managed agentic environments with least-privilege tool access and continuous red-teaming. The recommended approach is to ensure security maturity matches or slightly exceeds adoption maturity at every level to avoid both operational risk and regulatory non-compliance.
Related articles
OpenAI GPT-6 Astra Deploys Multi-Agent Parallel Processing, Increasing Local CPU Load and Security Risks
Early users of GPT-6 Astra have observed the model distributing complex tasks across multiple specialized agents that plan, solve, test code, verify results, and iterate after failures. This multi-agent approach enables faster handling of multi-step workflows compared to sequential chatbots. OpenAI states that Astra can control computers, operate browsers and applications, and install or test software, though it has not officially confirmed a native multi-agent architecture. Main computations run in the cloud, but agent tools can execute on user devices or corporate servers, leading to noticeable processor load when multiple agents compile code, launch browsers, run tests, and operate containers simultaneously. Corporate environments face added complexity as each agent requires virtual machines, sandboxes, internal data access, and careful environment cleanup. The increased autonomy has prompted OpenAI to strengthen monitoring of Astra actions and permission boundaries for subscribers of ChatGPT and enterprise clients.
Microsoft Copilot Can Surface Overshared Data Despite Permission Boundaries
Microsoft documentation states that Copilot only accesses data authorized for the signed-in user, yet default SharePoint and OneDrive sharing settings often grant broad access that the AI then respects literally. This creates accidental oversharing risks where Copilot retrieves documents shared too widely years earlier. Administrators can use Content Management Assessment and Data access governance reports, including the EEEU report covering the top 100 sites shared in the past 28 days, to identify problematic content. Two distinct controls exist: Restricted Access Control removes access entirely while Restricted Content Discovery hides items from Copilot and search without altering permissions. Sensitivity labels combined with encryption can exclude programmatic access for agents, though Microsoft does not guarantee outright blocking. Interaction logs stored in Microsoft Purview retain user prompts, Copilot responses, and citations to accessed documents, providing an audit trail for oversharing incidents.
Adaptive LLM Worm Uses Local Models to Craft Per-Target Exploits in Heterogeneous Networks
Researchers from the University of Toronto have published a preprint describing an adaptive computer worm driven by LLM agents that spreads across corporate networks by generating individualized attack strategies for each compromised system. Unlike traditional worms such as WannaCry that rely on fixed exploits, this worm maintains its own infrastructure by running local LLMs on infected GPU-equipped machines to analyze vulnerabilities and synthesize new attack vectors in real time. The system was tested in an isolated FakeCorp environment containing Linux, Windows, and IoT devices, successfully leveraging known real-world vulnerabilities to propagate over 48 hours and seven-day autonomous runs. Two core components power the worm: a GPU-hosted LLM component and a hierarchical agent framework with memory, reasoning graph, and tool modules that manage reconnaissance, exploitation, and payload deployment. The authors note that the approach creates an economic asymmetry favoring attackers because the worm parasitizes victim compute resources, eliminating the need for external C2 or commercial LLM services. They warn that adding adaptive reasoning to historical worms such as SQL Slammer, Conficker, or Stuxnet would significantly increase their resilience while remaining slower and noisier than classic self-propagating malware.
Building Secure On-Prem AI Assistants: How to Keep Corporate Data Inside Closed Contours
Many organizations hesitate to deploy AI assistants due to strict data protection rules that prohibit sending information to external clouds. The article explains how to implement AI models entirely within a company's own infrastructure, ranging from on-premise servers to fully offline laptops. It breaks down four deployment locations from public APIs to local devices and clarifies three distinct access levels: read, write, and execute. The author emphasizes that most business value comes from read-only access combined with human-in-the-loop controls for any irreversible actions. Practical recommendations include RAG over model size, quantization for local hardware, and maintaining immutable audit logs. The piece also warns that preparing clean knowledge bases often consumes more effort than the model itself.