[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fyk6hph6ZFyOW07GqsIQApC1ZuyAr3V2yyOGeCDcyj58":3},{"article":4,"iocs":45,"watch_terms":46,"solutions":47,"lesson":83},{"id":5,"title":6,"slug":7,"summary":8,"ai_summary":9,"brief":10,"url":11,"image_url":12,"published_at":13,"ingested_at":14,"relevance_score":15,"entities":16,"category_id":27,"category":28,"article_tags":32},"fc2590cd-ed2b-418c-9679-e6fe904e7e93","Managing open-weight model risk in enterprise deployments","managing-open-weight-model-risk-in-enterprise-deployments-e79b23","New whitepaper from Microsoft and NVIDIA helps enterprises manage open-weight AI model risks with a shared taxonomy, evaluations, and controls. The post Managing open-weight model risk in enterprise deployments appeared first on Microsoft Security Blog.","Microsoft and NVIDIA have released a whitepaper, 'Risk Management for Open-Weight Models in Enterprise,' to help organizations manage the risks associated with deploying open-weight AI models. The guidance introduces a shared risk taxonomy and an evaluation framework, emphasizing that risk is determined by the model, its use case, deployment, and surrounding controls, not just its openness. The paper aims to establish a common language for risk management across various enterprise teams and provides a structured approach to evaluating and mitigating AI-related risks.","Microsoft and NVIDIA release guidance for managing open-weight AI model risks.","https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fsecurity\u002Fblog\u002F2026\u002F10\u002F16\u002Fmanaging-open-weight-model-risk-in-enterprise-deployments\u002F","https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fsecurity\u002Fblog\u002Fwp-content\u002Fuploads\u002F2026\u002F10\u002FMSFT-NVIDIA-SCI.jpg","2026-10-16T16:00:00+00:00","2026-10-16T18:00:04.792606+00:00",7,[17,20,22,25],{"name":18,"type":19},"Microsoft","vendor",{"name":21,"type":19},"NVIDIA",{"name":23,"type":24},"AI","technology",{"name":26,"type":24},"open-weight models","839da5c1-3c34-47e2-9499-f7201640e3ac",{"id":27,"icon":29,"name":30,"slug":31},null,"AI Security","ai-security",[33,38,40],{"category":34},{"id":35,"icon":29,"name":36,"slug":37},"02371804-cf6d-4449-98de-f1a2d4d9b266","Tools","tools",{"category":39},{"id":27,"icon":29,"name":30,"slug":31},{"category":41},{"id":42,"icon":29,"name":43,"slug":44},"c5c77cdb-f7d7-4990-9436-c81dcbff1163","Policy","policy",[],[18,21],{"controls":48,"vendors":77},[49],{"slug":31,"name":50,"problem":51,"products":52},"AI application security","Prompt injection, data leaks through AI tools and unapproved AI use get caught.",[53,61,69],{"vendor_slug":54,"vendor_name":55,"vendor_logo":56,"product_slug":57,"product_name":58,"url":59,"description":60},"hiddenlayer","HiddenLayer","\u002Fvendor-logos\u002Fhiddenlayer.png","hiddenlayer-ai-security","HiddenLayer AI Security Platform","https:\u002F\u002Fhiddenlayer.com\u002F","Model scanning, AI red teaming and runtime protection for AI applications.",{"vendor_slug":62,"vendor_name":63,"vendor_logo":64,"product_slug":65,"product_name":66,"url":67,"description":68},"cisco","Cisco","\u002Fvendor-logos\u002Fcisco.png","ai-defense","Cisco AI Defense","https:\u002F\u002Fwww.cisco.com\u002Fsite\u002Fus\u002Fen\u002Fproducts\u002Fsecurity\u002Fai-defense\u002Findex.html","Discovery of AI use, testing of models and guardrails for AI applications.",{"vendor_slug":70,"vendor_name":71,"vendor_logo":72,"product_slug":73,"product_name":74,"url":75,"description":76},"palo-alto-networks","Palo Alto Networks","\u002Fvendor-logos\u002Fpalo-alto-networks.png","prisma-airs","Prisma AIRS","https:\u002F\u002Fwww.paloaltonetworks.com\u002Fai-security\u002Fprisma-airs","Security for AI applications, models and agents, including runtime protection against prompt injection.",[78],{"slug":79,"name":18,"logo_path":80,"roles":81},"microsoft","\u002Fvendor-logos\u002Fmicrosoft.png",[82],"reported_by",{"slug":84,"title":85,"body":86,"framework_refs":87,"framework_notes":96},"governing-open-weight-ai-models-requires-structured-enterprise-risk-frameworks","Governing Open-Weight AI Models Requires Structured Enterprise Risk Frameworks","As organizations increasingly adopt open-weight AI models, the absence of a shared risk taxonomy creates inconsistent governance and blind spots across enterprise teams. Unlike proprietary models, open-weight models can be fine-tuned and deployed without vendor guardrails, meaning risk is shaped as much by deployment context and surrounding controls as by the model itself. The Microsoft\u002FNVIDIA whitepaper highlights that organizations often lack structured evaluation frameworks, leaving AI deployments exposed to misuse, data leakage, and compliance failures. Without standardized risk language and assessment processes, security, legal, and business teams operate in silos, compounding exposure. Establishing governance frameworks before deployment — not after — is critical to responsible AI adoption.",[88,89,90,91,92,93,94,95],"NIST AI RMF (AI 100-1) — Govern, Map, Measure, Manage functions","NIST SP 800-53 Rev 5 — SA-11 (Developer Testing), RA-3 (Risk Assessment)","CIS Control 16 — Application Software Security","CIS Control 18 — Penetration Testing","EU AI Act — Article 9 (Risk Management System for High-Risk AI)","GDPR Article 25 — Data Protection by Design and by Default","ISO\u002FIEC 42001 — AI Management System Standard","MITRE ATLAS — Adversarial Threat Landscape for AI Systems",{}]