Military & security

arXiv:2607.26069v1 Announce Type: new Abstract: As AI systems are rapidly integrated into critical economic, governmental, and national security functions, the gap between AI adoption and AI security readiness continues to widen. This paper presents a prioritized agenda for advancing AI security, informed by structured interviews with leaders across industry, government, and civil society, and refined through a multi-sector expert workshop. Participants identified and ranked the highest-importan

Author Joyce Carol Oates came to the defense of Christopher Nolan’s “The Odyssey” after translator Emily Wilson wrote a viral review attacking the film. “Rather than disagreeing with interpretations of Homer in a collegial manner, this person, who has benefited enormously from Nolan’s film, speaks in the crude language of MAGA folks attacking someone with […]

The Enhancing K-12 Cybersecurity Act failed to pass in 2023. Sens. Mark Warner and Marsha Blackburn said their bill is urgently needed in 2026.

The agentic AI playing field was heavily tilted toward offense, so researchers began using red team agents to help teach their blue counterparts.

5 current and former government technology officials assess whether federal cyber defenses can keep up.

Agentic AI systems threaten cybersecurity as we know it. A new generation of cyber-capable models, including Anthropic’s Mythos and OpenAI’s GPT-5.6, can find and exploit vulnerabilities in computer systems far faster than human hackers. The technology is so powerful that it has spooked the U.S. government, which has moved to limit, or completely pause, the public release of these models. How well prepared is the Trump administration to secure the government’s computing resources? The rise of ag

The OpenAI CEO discussed a forthcoming artificial intelligence model with federal lawmakers as the cybersecurity debate over the advanced technology intensifies.

On 22 July 2026, the US administration announced a nuclear deal with Saudi Arabia. Under this deal, the US will provide support to the Golf country to establish a civilian nuclear programme. This is worrisome given statements by Saudi Arabia’s Crown Prince, that once Iran has nuclear weapons, Saudi Arabia would follow suit. Given the nuclear-induced tensions in the Middle East between Saudi Arabia and Iran and the scope of the agreement, this newly formed cooperation may put another nail in the

Dark Reading Confidential Episode 20: Expert Rich Mogull reflects on lessons cyber teams should pull from the OpenAI agent's attack on Hugging Face.

The initiative is intended to modernize and consolidate IT and data architectures at the Naval Surface Warfare Center Corona Division. The post Navy using LETHALITY Consortium to advance network consolidation initiative appeared first on DefenseScoop .

China has criticized the move, accusing the U.S. of protectionism.

The Federal Communications Commission (FCC) has banned imports of foreign-produced humanoid robotic devices and power inverters, citing “national security” as its primary concern. The devices were added to the FCC’s covered list, meaning the products posed an “unacceptable risk to the national security of the United States or the safety and security of United States...

Federated Learning (FL) is vulnerable to backdoor attacks because of its distributed nature in edge computing scenarios. Existing defense methods show limited efficacy as they overlook the deviations among benign local updates caused by statistical heterogeneity and the stealthiness of backdoor attacks. To tackle these issues, we propose FedDAB, a two-phase method that combines local contrastive regularization with alignment checking, to defend against backdoor attacks. In the first phase, FedDA

Canadian investment group to partner utility NextEra on data centre in Kentucky

The DRC’s case against Rwanda could mark an important moment in the legal development of responsibility for proxy warfare. The post DRC Takes Rwanda to the ICJ: Proxy Warfare, Layered Obligations, and the Future of State Responsibility appeared first on Just Security .

Das "KI-Update" liefert drei mal pro Woche eine Zusammenfassung der wichtigsten KI-Entwicklungen.

ChatGPT developer says activity by autonomous tool was not at severity or scale of what occurred at Hugging Face OpenAI has revealed that a cyber-attack carried out by a rogue AI agent had more than one victim. The ChatGPT developer said the agent – an autonomous tool able to carry out sequences of commands without human help – had located and used logins to access four other unnamed “publicly-available services” in addition to the US startup Hugging Face. Continue reading...

Open source communities have been flooded with AI-generated contributions. In defense, they have written contribution rules to regulate coding agents' behavior, spanning from a total ban, mandatory disclosure, to verification gates and human sign-offs. Yet, whether coding agents read and follow those rules, and behave in open source repositories, remains unknown. To estimate real-world rule compliance of coding agents, we curate 106 issues from 49 repositories containing AI contribution rules in

OpenAI has released Codex Security CLI, an open-source tool that automatically detects and fixes vulnerabilities in code repositories. Previously known internally as "Aardvark," the system has already helped fix more than 3,000 critical security flaws, according to OpenAI. It competes directly with Anthropic's Claude Security, as both AI companies race to match the growing automation of cyberattacks with AI-powered defense. The article OpenAI open-sources Codex Security CLI to help developers fi

Singapore's central bank is establishing an AI-Driven Cyber and Technology Risk Taskforce in collaboration with the country's leadingg commercial banks and tech companies.

In June, the U.S. government required Anthropic to restrict access to its two newest models by foreign nationals, citing national security and cybersecurity concerns. Unable to verify users’ nationality in real time, Anthropic temporarily withdrew access to the models for all users. A few weeks later, the Chinese company Moonshot AI released its Kimi K3 model, and then published its full model weights . Together, these episodes have reignited a familiar debate: How far can governments go in requ

This article presents a defense-in-depth approach for securing Model Context Protocol (MCP) deployments in production. It outlines four architectural control layers: safe execution, management infrastructure, outbound trust, and semantic integrity, arguing that production security requires enforcement beyond the gateway at the earliest trustworthy control points. By Nik Kale

Safety classifiers ("guards") are the dominant black-box defense for vision-language models, yet they judge an input's surface form, not its meaning: a harmful request re-encoded as set theory, formal logic, a rare language, code, or an image of text slips past a guard that would block it in plain language -- the decode gap. The natural fix is a guard-agnostic recover-and-decode amplifier that transcribes image content and restates encoded text into its plain payload before the guard, so any off

Companies need proactive and extended continuous threat exposure management to reduce cyber risk and cut the noise in security environments, says CyberCyte.

Research warns Washington should be alarmed at military implications of the sector’s rapid expansion

It's part of a national security strategy to kickstart domestic production of emerging technology.

Hollywood powerbroker Ari Emanuel rallied to the defense of his friend and business associate David Ellison, claiming that the state attorneys general trying to block Paramount’s Warner Bros. Discovery deal are threatening to “destroy” competition in the entertainment industry. Emanuel, who is CEO of TKO Group and former head of Endeavor, penned an op-ed published […]

Details have been released of an emergency call with hundreds of cyber-security experts after the ChatGPT hack of a tech company.

Boeing’s contract to replace two aging Air Force One planes, a program that is now four years behind schedule, lost the defense contractor $280 million in the second quarter of 2026, according to its latest earnings report released on Tuesday. The company inked a contract with the U.S. government in President Trump’s first term to...

The concept of “surveillance pricing” is just one part of a much larger problem and business model: corporations maximizing their profits by invading our privacy. The all-too-common business model is to systematically harvest, collate, and store as much of our personal data as possible, and then monetize it through use and sale. When it comes to surveillance pricing, that looks like corporations offering the same product to two different people at two different prices, based on harvested persona

National security agencies said the devices could create supply-chain vulnerabilities, threaten critical infrastructure and enable surveillance or remote manipulation.

Task Force Talon Synapse is slated to soon be fully operational, but some observers question its ultimate value. The post U.S. military forms first bilateral AI task force with United Arab Emirates appeared first on DefenseScoop .

The proposed legislation will authorize $10 million annually to address cyber threats in K-12 schools.

“It sounds good on paper,” a former senior defense official told Breaking Defense. “My biggest questions revolve around classified information sharing.”

Commerce Undersecretary William Kimmitt was at the Farnborough Airshow to develop a list of “high priority” potential sales, he told Breaking Defense.

Ein Brüsseler Leitfaden konkretisiert die Cyberresilienz-Verordnung. Hersteller und Open-Source-Projekte erhalten Rechtsgewissheit vor Start der Meldepflichten.

Fanning also gave The Break Out his take on the state of foreign military sales and the room for reform.

Microsoft bringt mit MAI-Cyber-1-Flash ein Security-Modell auf den Markt. Es soll nur halb so viel kosten wie Microsofts bisherige Security-Modellkombination.

An employee's email account had been compromised, allowing unauthorized access to "certain data," Bank of Baroda reported.

CORAS is a FedRAMP High-certified agentic-AI “decision maker” that has already been authorized for use at the Department of Defense. The post GSA inks agentic AI OneGov deal with CORAS appeared first on FedScoop .