xAI's Classified Access Stirs Debate, Cohere Teams with NVIDIA, and SAP Unveils Sovereign Cloud

Hi there,

Welcome to this week's edition of the AI Policy Brief, where we delve into the latest developments in AI regulations, safety standards, and compliance requirements worldwide. This week, we explore a range of stories that highlight the dynamic landscape of AI policy and its implications across various sectors. Senator Elizabeth Warren has raised concerns about the Pentagon's decision to allow xAI access to classified networks, citing potential national security risks. Meanwhile, Cohere has partnered with NVIDIA to enhance sovereign AI capabilities, focusing on secure, on-premise AI systems for regulated industries.

In other news, IntelePeer has achieved TX-RAMP Level 2 certification, confirming its compliance with stringent security standards for AI communication solutions. Humata Health has teamed up with AccuKnox to enhance security for its AI-driven healthcare platform, aiming to implement a Zero Trust CNAPP to comply with HIPAA requirements. Additionally, AEye, Inc. has joined forces with NVIDIA to enhance safety-certified AI solutions for automotive and infrastructure applications. Stay informed as we continue to cover these critical developments in AI policy and their global impact.

xAI's Access to Classified Networks Raises Concerns

Senator Elizabeth Warren has raised concerns about the Pentagon's decision to allow xAI access to classified networks, citing potential national security risks from its AI model, Grok. Read more

Cohere Partners with NVIDIA for Sovereign AI

Cohere has announced a collaboration with NVIDIA to enhance sovereign AI capabilities, focusing on secure, on-premise AI systems for regulated industries. This initiative addresses the demand for AI solutions that operate within national borders, particularly in finance, healthcare, and public services. Read more

IntelePeer Achieves TX-RAMP Level 2 Certification

IntelePeer has received TX-RAMP Level 2 certification, confirming its compliance with stringent security standards for AI communication solutions in healthcare and other regulated sectors. Read more

Humata Health Partners with AccuKnox for AI Security

Humata Health has teamed up with AccuKnox to enhance security measures for its AI-driven healthcare applications. The partnership aims to implement a Zero Trust CNAPP to comply with HIPAA regulations. Read more

University of Phoenix Study on AI in Higher Education

Researchers from the University of Phoenix have published a study on the use of generative AI in higher education, focusing on ethical and academic integrity issues. Read more

AEye Partners with NVIDIA for AI Safety Lab

AEye, Inc. has partnered with NVIDIA by joining the NVIDIA Halos AI Systems Inspection Lab to enhance safety-certified AI solutions for automotive and infrastructure applications. Read more

Microsoft May Sue OpenAI and Amazon Over Cloud Deal

Microsoft is considering legal action against OpenAI and Amazon due to a $50 billion deal that might violate its exclusive cloud agreement with OpenAI. The issue stems from OpenAI's partnership with Amazon Web Services, potentially conflicting with Microsoft's Azure platform. Read more

Germany to Expand AI Data Centres and Processing by 2030

The German government plans to significantly increase its AI data centre capacity and processing capabilities by 2030, as part of efforts to compete globally in the AI sector. Read more

Trump Administration Defends Blacklisting of Anthropic

The Trump administration has defended the Pentagon's decision to blacklist Anthropic, citing national security concerns in a court filing. Anthropic is contesting this action, arguing it infringes on its First Amendment rights. Read more

Patreon CEO Criticizes AI Companies' Fair Use Argument

At the SXSW conference, Patreon CEO Jack Conte criticized AI companies for using creators' work without compensation, challenging their fair use claims. Read more

AUVSI CEO Warns Congress on Chinese Robotics Risks

During a congressional testimony, AUVSI CEO Michael Robbins emphasized the strategic efforts by China to lead in robotics and AI, urging U.S. policy action to protect national security and economic interests. Read more

BMG Sues Anthropic Over AI Training with Lyrics

BMG Rights Management has filed a lawsuit against Anthropic for allegedly using copyrighted lyrics from artists like Bruno Mars and the Rolling Stones to train its AI chatbot. The lawsuit claims 493 instances of copyright infringement. Read more

Meta AI Agent Exposes Sensitive Data

An AI agent at Meta mistakenly exposed sensitive data to unauthorized employees for two hours, raising concerns about rogue AI agents within the company. Read more

SBS CyberSecurity Launches AI Peer Group for Financial Institutions

On March 20, 2026, SBS CyberSecurity introduced the SBS AI Peer Group to assist financial institutions in managing AI risks. The initiative offers a platform for banking professionals to explore AI use cases, governance, and regulatory updates. Read more

Meta Introduces AI for Content Enforcement

Meta has launched new AI systems to enhance content enforcement, aiming to reduce reliance on third-party vendors. These systems target content related to terrorism, child exploitation, and scams, with early tests showing improved accuracy and speed. Read more

Super Micro Co-Founder Resigns Amid Smuggling Charges

Yih-Shyan Liaw, co-founder of Super Micro Computer, has stepped down from the board following his arrest by the U.S. Justice Department for allegedly smuggling AI chips to China. The charges involve a scheme to reroute U.S.-made servers through Taiwan for repackaging and smuggling. Read more

Australian Government Sets Expectations for AI Infrastructure

The Australian Government has introduced a framework to guide data centers and AI infrastructure developers, focusing on community benefits and national interests. Read more

Russia Plans to Restrict Foreign AI Tools

The Russian government has proposed regulations to ban or restrict foreign AI tools, including ChatGPT and Claude, to ensure compliance with 'traditional Russian values' and mandate local data storage. Read more

Bernie Sanders Highlights AI Privacy Concerns in Video

Senator Bernie Sanders released a video to address privacy issues in the AI industry, using an AI chatbot named Claude to demonstrate his points. The video sparked criticism for potentially oversimplifying the complexities of data collection and privacy. Read more

SAP Launches Sovereign Cloud in France with Bleu

On March 19, 2026, SAP announced the launch of its Sovereign Cloud in France, developed in partnership with Bleu, to comply with French regulatory and security standards. Read more

Dutch Research Council Bans Hidden AI Prompts in Grants

The Dutch Research Council has prohibited the use of undisclosed AI prompts in grant applications to promote transparency and integrity. Read more

Anthropic Challenges Pentagon's National Security Claims

In a court filing, Anthropic disputes the Pentagon's claim that it poses a national security risk, arguing the government's concerns are based on misunderstandings. Read more

Pentagon Labels Anthropic a Supply-Chain Risk

The Pentagon has designated Anthropic as a supply-chain risk, a move criticized by U.S. Senator Elizabeth Warren as retaliation for the company's stance on military use of its AI technology. Read more

Palo Alto Networks Introduces Prisma Browser

On March 23, 2026, Palo Alto Networks announced the launch of the Prisma Browser, aimed at securing agentic AI workflows and addressing security risks like prompt injection attacks. Read more

Numonic Raises £250k for AI Content Tracking

London-based startup Numonic has secured £250,000 in pre-seed funding to enhance its platform for tracking AI-generated content, aiming to comply with the upcoming EU AI Act. Read more

White House Proposes National AI Framework

The White House has proposed a national AI policy to Congress, aiming to replace state regulations with a unified framework. The policy addresses child protection, energy management, and innovation, emphasizing the need for consistent legislation to maintain global competitiveness. Read more

Microsoft Unveils Security Features for AI at RSAC 2026

At the RSAC 2026 Conference, Microsoft introduced new security features for agentic AI, including the launch of Agent 365, to help IT and security teams manage AI agents and protect sensitive data. Read more

Unbound AI Introduces Security Platform for AI Coding Agents

Unbound AI has launched the Agent Access Security Broker (AASB), a platform aimed at improving governance and security for AI coding agents in enterprises. Read more

Blitz AI Launches AI Security Platform

Blitz AI, founded by former CrowdStrike leaders, has introduced a self-healing AI security platform that aims to automatically prevent and fix risks in real time, enhancing safety and governance for enterprises adopting AI technologies. Read more

Latent AI Leads Defense AI Working Group

Latent AI is heading a new Defense Working Group with the EDGE AI FOUNDATION to improve edge AI for defense operations. The group, chaired by Latent AI CTO Sek Chai, will tackle challenges in deploying AI at the tactical edge. Read more

Categories

Companies

Resources

xAI's Classified Access Stirs Debate, Cohere Teams with NVIDIA, and SAP Unveils Sovereign Cloud - AI Policy Brief #61

We hope you enjoyed this article.

More from: Regulation

More from: Data Protection & Privacy

Subscribe to AI Policy Brief