-
Trump Administration Veers from the Rule of Law in Singling Out Anthropic’s Latest Models
Transparency, fairness, clarity, and evidence-driven enforcement should be core components of AI governance. So long as AI policy decisions are driven by a few actors taking action behind closed doors in response to non-public reports, we will be far short of that standard.
-
-
ORNL-Developed Cybersecurity Framework Graduates from Lab to Electric Grid
A grid cybersecurity and verification framework developed at the Department of Energy’s Oak Ridge National Laboratory has been licensed by GridForge Energy Solutions. The startup plans to explore ways the technology could boost real-time visibility of grid behavior for energy projects, grid management companies and utilities.
-
-
Quantum Sensors Could Spot Hidden Damage in the Thousands of U.S. Bridges Rated “Structurally Deficient”
Damaged bridges is a national problem. The United States has more than 624,000 highway bridges. About 220,000 need major repair or replacement, and 41,677 are rated poor, also called structurally deficient.
-
-
Before SpaceX IPO, Investors in China Secretly Acquired Stakes
A businessman with ties to Chinese military contractors was among the overseas investors who acquired stakes in SpaceX while it was still a private company. SpaceX built its business off sensitive U.S. government work like making spy satellites for the Pentagon, so it is important to know which people in countries like China bought into the company, and how.
-
-
Why Anthropic Is Sounding the Alarm on the Next Generation of AI
Anthropic is urging its rivals to pursue an unprecedented regime of AI arms control. The explosive advances in the company’s technology illustrate why that effort may be imperative even as it will be difficult to achieve.
-
-
Nuclear in My Backyard: A Nebraska Utility Is Skirting the Public Backlash That Plagues Wind and Solar
Nationally, lawmakers in both parties have begun embracing nuclear power, as have everyday people. Across Nebraska, nuclear is getting a warm welcome in communities that typically oppose large-scale clean energy projects.
-
-
More States Restrict 3D-Printed Firearms
For decades, making an untraceable firearm required specialized tools, technical expertise and hours of work. Today, it can start with a downloaded file and a consumer-grade 3D printer. At least six states this year have outlawed or imposed regulations on such guns.
-
-
Going Dark: Why Dismantling America’s Ocean Sensors Is a Security Risk
Starting this month, the National Science Foundation (NSF) began pulling more than 900 deep-sea instruments out of the water, dismantling most of the Ocean Observatories Initiative (OOI), a network of moorings and seafloor sensors that streams continuous, real-time data on the physical, chemical, and biological state of the Atlantic and Pacific. The network was built to run for 25 years. It will not make it to ten. This is a national security story as much as a science story.
-
-
For First Time, Americans Are Getting More of Their Electricity from Solar Than Coal
Solar energy just provided more electricity in the United States than coal for the first time on record — marking a milestone for the rise of renewables in America. Solar provides more than twice the share of electricity it did five years ago.
-
-
AI-Generated Zero-Day Vulnerabilities: Implications for Future Warfighting
Cybersecurity has long been viewed as a support function, largely focused on protecting military computer networks and information systems. This paradigm is now obsolete. AI is now reshaping the character of warfare itself, turning cybersecurity into a foundational layer of combat power.
-
-
From Exporting Spyware to Surveilling Activists – How Democracies Became the New Digital Authoritarians
“Digital authoritarianism” refers to governments using technology for surveillance and censorship to repress dissent. China remains the master practitioner, but democracies, too — in particular, India and Israel — are beginning to repress their citizens with the same tools and export them abroad.
-
-
Will Generative AI Fundamentally Change Terrorist Threats?
Does generative artificial intelligence (Gen AI) represent a genuine transformation in terrorist capability, or are its risks being overstated within contemporary security discourse? Gen AI tools, particularly large language models, can improve the efficiency, accessibility, and scale of certain terrorist activities, but there remains limited evidence that they fundamentally alter the nature of terrorism or significantly enhance operational capability.
-
-
From Oversight to Coercion: How Authoritarian Governments Are Twisting AI Safety to Get Tech Companies to Fall in Line
Authoritarian governance of technology often delegitimizes intended protections, poisoning any external regulation, and even voluntary self-regulation that deviates from the regime’s goals or values. Following the authoritarian playbook, the Trump administration has argued that AI safety standards and user restrictions are ideological impositions rather than sound engineering decisions. In the “Preventing Woke AI” executive order of 23 July 2025, the administration attached the “woke” label to basic ethics protections, making those protections politically costly to maintain.
-
-
The Challenges of Collaborating with China on Space Research and Technology
China advances its space research and development to the detriment of other countries’ space goals. China’s path to space power has left a trail of disappointed partner nations in its wake, producing lopsided outcomes in which partners assume financial and political risk while ceding control over infrastructure, data, and dual use technologies.
-
-
Not Regulation. Not Even Oversight. Trump’s AI Order Won’t Be Enough
After a bruising tussle inside his administration, President Donald Trump has gone for a decidedly non-committal approach to AI safety and security that’s unlikely to survive sustained contact with reality.
-
The long view
Will Generative AI Fundamentally Change Terrorist Threats?
Does generative artificial intelligence (Gen AI) represent a genuine transformation in terrorist capability, or are its risks being overstated within contemporary security discourse? Gen AI tools, particularly large language models, can improve the efficiency, accessibility, and scale of certain terrorist activities, but there remains limited evidence that they fundamentally alter the nature of terrorism or significantly enhance operational capability.
AI-Generated Zero-Day Vulnerabilities: Implications for Future Warfighting
Cybersecurity has long been viewed as a support function, largely focused on protecting military computer networks and information systems. This paradigm is now obsolete. AI is now reshaping the character of warfare itself, turning cybersecurity into a foundational layer of combat power.
More States Restrict 3D-Printed Firearms
For decades, making an untraceable firearm required specialized tools, technical expertise and hours of work. Today, it can start with a downloaded file and a consumer-grade 3D printer. At least six states this year have outlawed or imposed regulations on such guns.