Archive

artificial intelligence

Q&A: From Capability to Trust: AI Agents & Agency in the Internet Ecosystem

Q&A: From Capability to Trust: AI Agents & Agency in the Internet Ecosystem

As AI agents rapidly reshape how the internet functions, how can we preserve trust and stability in our digital ecosystem? Today’s report examined core challenges presented by the proliferation of agents across our systems, and offered recommendations for policymakers, industry leaders, and more.

agency, agentic, agents, AI, artificial intelligence, attribution, authorization, identity, intellectual property, stability, trust

April 30, 2026

AI Agents & Agency in the Internet Ecosystem

Report, White Paper

AI Agents & Agency in the Internet Ecosystem

AI agents are rapidly reshaping how the internet functions. Software systems operate, interact, and execute tasks autonomously – marking a structural shift in our digital ecosystem. A new report presents three core insights into the AI agent landscape, supporting activities by policymakers, industry leaders, and standards bodies to enhance trust and stability.

agency, agentic, agents, AI, artificial intelligence, attribution, authorization, identity, intellectual property, stability, trust

April 30, 2026

What Anthropic’s Mythos Preview Tells Us About AI Loss of Control Risk

What Anthropic’s Mythos Preview Tells Us About AI Loss of Control Risk

The release of Claude Mythos Preview has made cyber headlines. But instead of viewing it through the lens of misuse, policymakers should actually be approaching Mythos through the lens of AI Loss of Control, argue Ritika Verma and Mariami Tkeshelashvili.

AI, artificial intelligence, cyber, loss of control, Mythos, system card

April 29, 2026

Design Obligation or Third-Party Liability? The White House Framework and AI’s Grey Area

Design Obligation or Third-Party Liability? The White House Framework and AI’s Grey Area

The White House’s newly-released National Policy Framework for Artificial Intelligence addresses the boundary between federal and state AI regulation, including a recommendation that states should not hold developers liable for third parties’ unlawful use of their models. Gabrielle Tran examines what this means for the emerging legal landscape around AI harm.

AI, artificial intelligence, chatbot, Executive Order, generative AI, liability, Section 230, third party

April 13, 2026

Meet the Andrew Carnegie AI-Nuclear Policy Accelerator Inaugural Cohort

Meet the Andrew Carnegie AI-Nuclear Policy Accelerator Inaugural Cohort

IST introduces the 26 members of the inaugural Andrew Carnegie AI-Nuclear Policy Accelerator cohort, a program that will equip mid-career security practitioners with the tools to confront the intersection between AI and nuclear issues.

AI, announcement, arms control, artificial intelligence, fellows, ICBM, modernization, nuclear, nuclear weapons

April 9, 2026

Meet the Religious Voices and Responsible AI Steering Committee

Meet the Religious Voices and Responsible AI Steering Committee

Thirteen computer science, philosophy, theology, and public policy experts have joined IST and AI and Faith to provide strategic guidance on the Religious Voices and Responsible AI initiative.

AI, artificial intelligence, Christianity, faith, Islam, religion

March 19, 2026

Reflecting on the Munich Cybersecurity Conference Roller Coaster: Digital Sovereignty, Decoupling, and the Risks and Opportunities of AI

Reflecting on the Munich Cybersecurity Conference Roller Coaster: Digital Sovereignty, Decoupling, and the Risks and Opportunities of AI

IST Chief Strategy Officer Megan Stifel visited Germany in February to join cybersecurity experts from around the globe at the Munich Cyber Security Conference. From digital sovereignty to the risks and opportunities of AI, she reflects on her experience for the NatSpecs blog.

artificial intelligence, cybersecurity, decoupling, digital sovereignty, Munich, transatlantic

March 5, 2026

Lessons from Moltbook: When Agents Talk to Agents

Lessons from Moltbook: When Agents Talk to Agents

Moltbook, a Reddit-like forum meant solely for AI agents, had already garnered over 1.5 million users before research suggested that a multi-agent environment with weak cybersecurity guardrails could be a hotbed for scaling fraud and influence operations. For the IST blog, Gabrielle Tran outlines what this means for the cyber-enabled manipulation of tomorrow.

AI, AI agents, artificial intelligence, fraud, malware, phishing, prompt injection, trust

February 26, 2026

A Changing Export Control Landscape: H200 Exports, Remote Access Rules, and What Comes Next

A Changing Export Control Landscape: H200 Exports, Remote Access Rules, and What Comes Next

January 2026 was, in many ways, an inflection point for export controls. From the resumption of Nvidia H200 sales to China to congressional discussion on restricting remote cloud access, IST's AI Chip Export Control Initiative team unpacks what these developments mean for the future.

AI, artificial intelligence, Bureau of Industry and Security, China, chip, export control, NVIDIA, semiconductor

February 24, 2026

Q&A: An AI Loss of Control “Warning Shot”

Q&A: An AI Loss of Control “Warning Shot”

How might you lose control of a system as complex as a frontier AI model? In this month's newsletter, we speak to the authors of IST's newest report diving into the risk of AI Loss of Control – that is, models diverging from authorized constraints, to the extent that a human operator can no longer prevent or constrain undesired outcomes - and presenting a framework for mitigation.

AI, AI Loss of Control, artificial intelligence, deception, model drift, risk reduction, scheming

February 19, 2026

Something Mysterious Is Happening

Something Mysterious Is Happening

In a blog reflecting on her co-authored report, AI Loss of Control Risks, Mariami Tkeshelashvili writes, "There is a difference between embracing mystery as a driver of discovery and being incurious about risks that are already becoming visible. Monitoring AI behaviours that we cannot fully explain is not pessimism—it is responsibility. It is what keeps us safe."

AI, artificial intelligence, deception, Indications and Warning, loss of control, manipulation, monitoring, risk reduction, scheming, strategic foresight

February 19, 2026

AI Loss of Control Risk: Indications & Warning

AI Loss of Control Risk: Indications & Warning

Though technologists and policymakers alike are eager to address AI Loss of Control–a state in which an AI system diverges from authorized constraints–there are significant gaps in the ways stakeholders understand, anticipate, and perceive this risk. "AI Loss of Control Risk" proposes applying the Indications & Warning (I&W) methodology used by the intelligence community to monitor this risk.

AI, AI Loss of Control, artificial intelligence, deception, model drift, risk reduction, scheming

February 19, 2026

IST Launches the Andrew Carnegie AI–Nuclear Policy Accelerator

IST Launches the Andrew Carnegie AI–Nuclear Policy Accelerator

The Institute for Security and Technology (IST) is pleased to announce the launch of the Andrew Carnegie AI–Nuclear Policy Accelerator, a new, practitioner-focused initiative designed to strengthen decision-making at the intersection of artificial intelligence (AI) and nuclear policy.

AI, announcement, artificial intelligence, fellowship, NC3, nuclear policy, nuclear systems

February 10, 2026

Autonomous Agents, Human Consequences: Key Insights from IST’s Workshop on AI Agents & Agency in the Internet Ecosystem

Autonomous Agents, Human Consequences: Key Insights from IST’s Workshop on AI Agents & Agency in the Internet Ecosystem

AI agents are shaping how decisions are made, how systems behave, and how humans navigate the digital world. To better understand the implications of this shift, IST hosted a closed-door workshop to explore the potential effects of AI agents on human agency.

Agentic AI, AI, AI agents, artificial intelligence

December 19, 2025

AI and NC3 Initiative Enters Phase 2: Bridging Perspectives on Risks and Opportunities

AI and NC3 Initiative Enters Phase 2: Bridging Perspectives on Risks and Opportunities

IST, with support from Longview Philanthropy, is entering phase 2 of our efforts on the integration of artificial intelligence into nuclear command, control, and communications. In its next phase, the AI and NC3 initiative will establish an executive committee and 4 working groups, driving further research on competitive AI dynamics, global perspectives on AI-NC3, AI technical development and trajectories, and AI norms and governance.

AI, artificial intelligence, competitive AI, governance, NC3, norms, nuclear, strategic stability

December 11, 2025

Giving Tuesday 2025: IST’s Impact

Giving Tuesday 2025: IST’s Impact

Every day, the work we do at IST makes a difference, from driving Congress to consider new approaches to intractable problems, to bringing national security leaders together to ensure they consider the existential threat of AI, to arming K-12 schools with the tools they need to defend against ransomware. As a 501(c)(3) nonprofit, this is only possible because of the generous support of our partners and donors.

AI, artificial intelligence, critical infrastructure, cybersecurity, Giving Tuesday, national security, Ransomware

December 2, 2025