CISO
Enterprise Architect
Compliance Officer
Industry relevance
Financial Services
Healthcare
Government
Energy
DECEMBER 16, 2025
NIST's Cybersecurity Framework now has an AI extension — AI systems are formally in scope for cybersecurity controls, not outside them.

NIST NCCoE published the preliminary draft of NIST IR 8596, Cybersecurity Framework Profile for Artificial Intelligence (Cyber AI Profile), in December 2025. The profile extends the NIST Cybersecurity Framework to AI systems across three focus areas: securing AI system components, conducting AI-enabled cyber defense, and governing AI across the enterprise. Public comment closed January 30, 2026.
GOVERNANCE IMPLICATION
The NIST Cybersecurity Framework Profile for Artificial Intelligence establishes that AI systems are within the CSF scope — not a separate domain governed by different frameworks. For regulated organizations that use the NIST CSF as their compliance baseline, the Cyber AI Profile creates an expectation that AI deployments are assessed against the same Identify, Protect, Detect, Respond, and Recover functions as any other technology system. Organizations that have treated AI governance as separate from their CSF program now have a documented gap between what NIST expects and what their program covers.
SCENARIO
A regional bank uses the NIST Cybersecurity Framework as its primary compliance baseline submitted to OCC examiners each cycle. In December 2025, NIST publishes the preliminary Cyber AI Profile extending CSF to AI systems. By the time the bank's next examination arrives in Q3 2026, the examiner references the profile and asks which CSF functions have been applied to the bank's Copilot and agent deployments. The bank's CSF program has no AI-specific controls mapped to any CSF function.
THE GOVERNANCE QUESTION
The NIST Cyber AI Profile establishes that AI systems require cybersecurity controls as first-class requirements — not add-ons addressed after deployment. If your organization uses the NIST Cybersecurity Framework as a compliance or audit baseline, the Cyber AI Profile creates the expectation that your AI deployment inherits that framework. Has your security team assessed which CSF controls apply to your agent deployments, or is AI still being treated as outside the framework boundary?
CONTROL GAP
Most NIST CSF implementations were built before AI agents existed as an enterprise risk category. The Cyber AI Profile creates a mapping expectation that existing CSF programs have not fulfilled — AI is not in the asset inventory, not in the risk register, and not in the incident response playbook.
REGULATORY RELEVANCE
NIST Ai RMF
FFIEC
OCC
FINRA
SEC Cyber
PRIMARY SOURCE
NIST releases preliminary draft of Cyber AI Profile
NIST NCCoE
Read the primary source ->(opens in new tab)CONTINUE READING

OCTOBER 16, 2026
ComplianceOn October 16, 2026, Microsoft and NVIDIA published Risk Management for Open-Weight Models in Enterprise, the first in a planned series of joint resources. The paper introduces a shared taxonomy of five risk categories: Cybersecurity, Data Protection, Content, Human-AI Interaction, and Bias. Microsoft says enterprises should connect relevant risks to evaluations, mitigations, and accountable owners, and the paper draws on ISO standards and the NIST AI Risk Management Framework. Microsoft states that weight openness alone does not determine risk, that responsibility for safeguards depends on the service and deployment configuration, and that evaluating a model alone does not establish the safety of an agent or application. The first paper focuses on the model layer.

SEPTEMBER 1, 2026
ComplianceOn September 1, 2026, Microsoft published its third annual Responsible AI Transparency Report. Chief Responsible AI Officer Natasha Crampton wrote that Microsoft re-engineered its Responsible AI Standard so that it is structured by technology stack component, covering models, platform services, and applications, and by the role Microsoft plays in each. Microsoft says its technical risk management now focuses on controls such as agent identities, tool permissions, and monitoring of actions. Tools named include an AI Red Teaming Agent, agent evaluators, RAMPART, ASSERT, and the Agent Control Specification. Microsoft states it is one of the few companies certified against ISO 42001 across a broad portfolio.

JULY 27, 2026
ComplianceRegulation (EU) 2026/1744, the Digital Omnibus on AI, is dated 8 July 2026. Its Official Journal header shows publication on 24 July 2026, and it enters into force on the third day following publication, 27 July 2026. Recital 40 sets the date of application of the high-risk requirements for AI systems classified under Article 6(2) and Annex III at 2 December 2027. This note covers only the dates confirmed against the Official Journal text and leaves other changes in the regulation to a separate review.