AI & Technology

NATO's current intelligence sharing policies are severely outdated, creating barriers to rapid dissemination of commercially-generated and AI-processed intelligence crucial for military decision-making. A top NATO intelligence official warns that the alliance relies on ad-hoc workarounds, jeopardizing operational speed in a contested global environment.

A groundbreaking investigation into 1 million exposed AI services reveals that AI infrastructure is significantly more vulnerable, exposed, and misconfigured than any other software previously studied. The rapid pace of AI adoption, particularly in self-hosted LLM environments, is leading businesses to prioritize speed over robust security measures. This critical oversight is creating massive new attack surfaces and exposing sensitive data to malicious actors.

Cisco has announced its intent to acquire Astrix Security, a startup specializing in securing non-human identities (NHIs) such as API keys and service accounts. This strategic move aims to extend zero trust principles to the rapidly expanding "agentic workforce" of AI agents and machine identities, addressing critical emerging risks within the enterprise attack surface.

The Pentagon has secured groundbreaking deals with seven major tech companies, including Google, Microsoft, and OpenAI, to integrate their advanced artificial intelligence capabilities into classified military networks. This strategic move aims to significantly augment warfighter decision-making and streamline operations in complex environments, signaling a major acceleration in the Department of Defense's AI adoption. The partnerships highlight the growing reliance on private sector innovation for national security technologies.

Cisco has launched an open-source Model Provenance Kit designed to bolster the integrity and security of third-party AI models used by organizations. This new Python-based toolkit aims to mitigate significant risks such as model poisoning, inherent biases, and unverified claims from model developers by providing a robust 'fingerprint' for tracing AI model lineage.

The U.S. Navy has awarded a significant contract to AI firm Domino Data Lab, accelerating its capabilities to counter Iranian mine threats in the critical Strait of Hormuz. This initiative, part of Project AMMO, aims to drastically reduce the time needed for underwater drones to identify new mine types, from months to mere days. The move comes amid heightened tensions and concerns over global shipping lane security.

As the U.S. military rapidly integrates increasingly autonomous AI into warfighting, a critical flaw in current human control paradigms has been identified. Simply requiring human approval for lethal force at the final stage is deemed insufficient, risking algorithms making impactful decisions that unacceptably constrain human choice. Experts are now advocating for "Synthesized Command & Control," a novel approach that embeds human preferences and commander's intent directly into AI algorithms from the outset.

The U.S. Army is exploring significantly increased autonomy for AI "agents" in cyber defense, particularly during wartime, following a recent wargame with senior tech industry executives. This shift includes developing a critical "risk continuum" policy to define when agentic AI can operate with minimal human oversight against fast-evolving cyber threats.

The Department of Defense has finalized agreements with seven major artificial intelligence companies, including OpenAI and Google, to integrate their systems into classified Pentagon networks. This significant push to establish the U.S. military as an AI-first fighting force notably excludes Anthropic, which was previously designated a supply-chain risk due to disputes over data access for autonomous weapons and surveillance.

In a significant move to accelerate AI integration into national security, the Pentagon has approved seven leading artificial intelligence developers to deploy tools within its highly classified Impact Level 6 and 7 networks. This broad strategic initiative aims to streamline data synthesis, enhance warfighter decision-making, and prevent vendor lock by fostering a diverse AI ecosystem. The integration underscores a presidential mandate to solidify American AI leadership as indispensable to defense capabilities.

The Pentagon has announced a significant step in its AI-first strategy, authorizing seven major tech firms to deploy their artificial intelligence models directly onto its highly classified networks, including Impact Level 6 and the semi-official Impact Level 7. This strategic decision accelerates the integration of commercial frontier AI capabilities, aiming to enhance data synthesis and elevate warfighter decision-making in complex operational environments.

The former head of the Defense Department’s Office of Net Assessment (ONA), James Baker, has joined leading AI company Anthropic as a strategist-in-residence. Baker, who previously advised defense secretaries on emerging technology's long-term effects on national security, will now lead analysis on how artificial intelligence is impacting U.S. institutions and competition with China. This move signifies a critical convergence of high-level national security strategy and cutting-edge AI development.