The Exchange Daily

Federal leaders face accelerating mandates on AI governance, cyber defense posture, and IT modernization. Here is the verified intelligence you need today. CAISI/NIST Expands Frontier AI Pre-Deployment Vetting with Google DeepMind, Microsoft, and xAI The Center for AI Standards and Innovation at NIST has broadened its evaluation partnerships with leading frontier model developers. These agreements...

CAISI Delivers Implementation Resources for Updated AI Benchmarks The Center for AI Standards and Innovation is releasing practical support materials to help agencies apply real-world performance testing and supply-chain risk scoring in current high-impact AI procurements. These resources translate the May 8 benchmarking guidance into actionable steps that procurement teams and technical evaluator...

Executive Summary System integrators and service providers received the clearest signal yet this week on how the federal government intends to procure and deploy high-impact artificial intelligence systems. Coordinated announcements on May 8 across NIST, GSA, FedRAMP, OMB, and the Department of Defense transformed the March 2026 GSA-NIST partnership from strategic memorandum into operational procu...

NIST Releases Updated Benchmarking Guidance for High-Impact AI The Center for AI Standards and Innovation has issued new benchmarking guidance for high-impact AI systems. Agencies must now incorporate real-world performance testing and supply-chain risk scoring into all procurement decisions. GSA Accelerates AI-Specific Contract Vehicles via USAi GSA is fast-tracking new AI contract vehicles throu...

CAISI Secures Pre-Deployment Reviews with Microsoft, Google DeepMind, and xAI The Center for AI Standards and Innovation has agreements with Microsoft, Google DeepMind, and xAI for national security evaluations of frontier models prior to public release. Reviews target cybersecurity, biosecurity, and chemical weapons risks to strengthen federal preparedness. GSA and NIST Deepen AI Evaluation Partn...

Navy Reports Measurable Efficiency Gains from AI Training Tools The Navy is documenting significant efficiency improvements from AI-powered training platforms and operational workflows. Leaders report faster planning cycles, reduced administrative overhead, and enhanced readiness metrics across fleet and shore-based commands. FedRAMP Advances Continuous Authorization for AI Services FedRAMP is rol...

DoD Secures Landmark AI Partnerships for Classified Networks The Department of Defense has entered agreements with SpaceX, OpenAI, Google, NVIDIA, Microsoft, AWS, Reflection, and Oracle to bring frontier AI capabilities onto IL6 and IL7 classified networks. The initiative strengthens warfighting, intelligence, and operational superiority while mitigating vendor lock-in and supply-chain risks. CISA...

GSA and NIST Launch Joint AI Evaluation Partnership for Federal Procurement The General Services Administration and National Institute of Standards and Technology formalized a partnership to embed rigorous AI evaluation science into federal procurement, strengthening testing for the USAi platform and accelerating safe adoption. GAO Highlights Persistent AI Acquisition Data Gaps Across Agencies The...

House Launches Joint Probe into PRC AI Model Risks House Homeland Security Chairman Andrew Garbarino and Select Committee on China Chairman John Moolenaar announced a formal investigation into national security risks from PRC-developed AI models, including potential distillation of U.S. frontier systems. Letters have been sent to key companies seeking exposure details. FDA Seeks Input on AI Pilot...

NIST Delivers AI RMF Profile for Critical Infrastructure Trustworthiness NIST’s new concept note for an AI Risk Management Framework Profile gives critical infrastructure operators a repeatable, lifecycle approach to map, measure, manage, and govern AI risks in IT, OT, and ICS environments. The profile directly addresses safety, security, reliability, and resilience demands as AI moves into high s...

AI Governance & Policy: White House National AI Legislative Framework On March 20, 2026, the White House released its National AI Legislative Framework. The six-objective plan protects children and parents, drives innovation, and preempts conflicting state AI laws to preserve U.S. competitiveness while respecting federalism. It is the clearest signal yet that Washington intends to lead the global...

White House OSTP Issues Guidance on Countering Foreign Industrial-Scale AI Model Distillation and Theft The White House Office of Science and Technology Policy has released new guidance addressing sophisticated foreign campaigns targeting U.S. frontier AI models. The NSTM-4 memo highlights deliberate efforts, primarily from China, that use proxy accounts and advanced jailbreaking techniques to ext...

GSA-NIST Partnership Strengthens AI Evaluation for Federal Procurement The General Services Administration and NIST signed a memorandum of understanding to enhance AI model testing and benchmarking through the USAi secure platform. Agencies gain consistent, workflow-ready evaluation tools before deployment. NIST Advances Trustworthy AI Profile for Critical Infrastructure NIST published the concept...

Bipartisan Commission to Shape AI Economic Policy Congress introduced the Economy of the Future Commission Act of 2026 to study artificial intelligence’s impact on jobs, education, federal adoption, and U.S. competitiveness. The bicameral commission will deliver actionable recommendations on workforce reskilling and long-term growth. DOE FY 2027 Budget Delivers $1.2 Billion for AI Supercomputing T...

White House Invokes Defense Production Act for AI Baseload Power The White House is using the Defense Production Act to secure coal supply chains and baseload power for AI energy demands and national defense. This policy signal elevates power availability to mission-critical status. DOE Issues RFI for AI Infrastructure on Federal Lands The Department of Energy seeks input on sixteen potential site...

White House National AI Policy Framework: Agencies Already Executing the Seven Pillars The March 20, 2026 White House National Policy Framework for Artificial Intelligence is no longer a document – it is an operating system. This week the Department of Education finalized its supplemental AI priority for grants (effective May 13), the Air Force advanced its 4,700-acre Alaska AI data-center plans u...

White House Grants Agencies Controlled Access to Anthropic Mythos Model OMB is establishing protections so major federal agencies can begin using Anthropic’s unreleased Mythos preview under Project Glasswing for defensive missions. OMB Releases 2025 AI Use Case Inventory – 3,611 Cases and 105 Percent Growth The official repository now documents dramatic expansion across fifty-six agencies, with HH...

DIA Activates Permanent Digital Modernization Accelerator to Scale AI Enterprise-Wide The Defense Intelligence Agency has stood up the hub-and-spoke Digital Modernization Accelerator to centralize governance and funding while empowering mission teams to deliver rapid AI use cases. Built on Task Force Sabre lessons, the permanent structure already includes ChatDIA and is advancing agentic AI inside...

DOE and LLNL Launch Mjölnir AI Testbed for Energy-Sector AI Cybersecurity CESER and Lawrence Livermore National Laboratory activated the Mjölnir AI Testbed on April 14. The platform lets stakeholders upload energy-sector AI models and run adversarial tests to expose vulnerabilities in grid management, operations planning, and forecasting applications. DOT Scales AI-CHAT Chatbot to 5,000+ Federal U...

GAO Urges Systematic Lessons Learned from Federal AI Acquisitions Federal agencies more than doubled AI use from 2023 to 2024, yet GAO-26-107859 (released April 13) finds they are not systematically collecting lessons learned. The report recommends policy updates at DOD, DHS, GSA, and VA to feed best practices into the GSA repository. NIST Advances Trustworthy AI Standards for Critical Infrastruct...

Education Department Finalizes AI Supplemental Priority for Federal Education Programs The Department of Education published its final supplemental priority and definitions on advancing artificial intelligence in education. The guidance directly shapes grant programs, curriculum standards, and federal education technology investments. Agencies and contractors supporting education IT should align p...

GAO Urges OMB to Close Privacy Gaps in Federal AI Guidance The Government Accountability Office reports that OMB has yet to fully address privacy vulnerabilities in current AI memorandums. Agencies remain exposed on sensitive data handling. CIOs and CISOs must review the findings and advocate for updated guidance before scaling deployments. Federal Reserve Data Shows Sharp Rise in AI Adoption Nati...

Data Readiness Emerges as the Real Bottleneck for Federal AI Adoption Federal agencies are spending up to eighty percent of AI project time on data preparation rather than model deployment. The White House AI Action Plan set clear priorities for innovation and infrastructure, yet without governed, ready data pipelines, pilots never reach scale. CIOs must audit data estates immediately and embed au...

AI risk profiles, workforce readiness, and privacy gaps dominate this edition. NIST Launches Concept Note for AI RMF Profile on Trustworthy AI in Critical Infrastructure NIST released the concept note on April 7 for a new profile under the AI Risk Management Framework focused on trustworthy AI in critical infrastructure sectors. The profile provides specific risk management practices for operators...

White House FY2027 Budget Requests Record $75.7 Billion for Civilian IT with Major AI Focus The White House released its Fiscal Year 2027 civilian IT budget request totaling $75.7 billion, an $8 billion increase. The funding prioritizes AI adoption, cloud infrastructure modernization, and mission-critical upgrades at the VA, Treasury, and DOJ. GSA-NIST Partnership Strengthens AI Evaluation and Tes...