Archive

risk reduction

Meet the Executive Committee Members, Working Group Co-Chairs, and Working Group Experts Contributing to IST’s AI-NC3 Initiative

Leading academics, technologists, and operators have joined the Artificial Intelligence and Nuclear Command, Control, and Communications Initiative to sit on its Executive Committee, chair the four Working Groups, and offer expert guidance as members of the Working Groups.

AI, artificial intelligence, command authority, decision support, early warning, governance, NC3, norms, nuclear, risk reduction

June 1, 2026

UnDisruptable27 Newsletter: “The threat comes from cyber, but the solutions come from engineering. It’s not shields up. It’s connections down.”

In the latest edition of the UnDisruptable27 newsletter, the team reflects on Josh's testimony before Congress, surfacing some of the most surprising takeaways, key insights from his fellow witnesses, and what the community can do next.

cascading failures, congress, critical infrastructure, cyber-informed engineering, cybersecurity, risk reduction, water, water utilities

May 28, 2026

Crossed Wires and Rising Escalations: IST Hosts Workshop on Reinvigorating Nuclear Crisis Communications

On February 18, 2026, the Institute for Security and Technology’s Nuclear Policy team convened for a workshop and scenario-based exercise titled “Reinvigorating Nuclear Crisis Communications in an Era of Emerging Technologies” with leading diplomats, policy experts, and technologists in Geneva, Switzerland.

arms control, crisis communications, escalation, mesh networks, New START, nuclear, risk reduction

March 9, 2026

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

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

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

New Video Narrated by Rose Gottemoeller Spotlights “CATALINK”: A Critical Blueprint for Secure, Multilateral Crisis Communications

Narrated by Rose Gottemoeller, former NATO Deputy Secretary General and U.S. Undersecretary of State for Arms Control and International Security, a new three-minute animated film live today introduces CATALINK. This groundbreaking initiative is a blueprint for a dedicated, last-resort communication system designed to be the ultimate backstop to prevent or cease nuclear war.

arms race, CATALINK, crisis communications, hotlines, mesh network, nuclear, Puck, risk reduction

December 17, 2025

IST Statement on President Trump’s AI Action Plan and Related Executive Orders

In February, IST submitted comments in response to the Office of Science and Technology Policy’s request for information on the development of an AI Action Plan. We were heartened to see many of the key issues we raised reflected in the Trump Administration’s recent AI executive orders.

AI, artificial intelligence, chips, compute, distributed energy, energy security, export control, national security, risk reduction

July 25, 2025

Responding to the Unknown: Simulating AI-Driven Crises

In April, Mariami Tkeshelashvili and Jennifer Tang hosted an AI crisis simulation exercise at the Johns Hopkins SAIS Emerging Technologies Symposium. The hands-on exercise asked participants to explore challenges at the intersection of AI and national security, and used IST’s research to set the stage before diving into immersive crisis scenarios.

AI, cyber, national security, risk reduction, risk-mitigation, scenarios, workshop

May 13, 2025

Navigating AI Compliance, Part 2: Risk Mitigation Strategies for Safeguarding Against Future Failures

"Navigating AI Compliance, Part 2: Risk Mitigation Strategies for Safeguarding Against Future Failures" presents 39 risk mitigation strategies co-created by a multistakeholder working group of experts that aim to avoid institutional, procedural, and performance failures of AI systems.

AI, artificial intelligence, compliance failure, infrastructure protection, institutional failures, performance failures, procedural failures, risk reduction

March 20, 2025

Managing Misuse Risk for Dual-Use Foundation Models: IST Submits Comments to a NIST Request for Information

Last week, Institute for Security and Technology (IST) submitted a response to NIST's Request for Comments on the U.S. Artificial Intelligence Safety Institute's draft guidelines for identifying and mitigating the risks to public safety and national security present across the AI lifecycle.

AI, artificial intelligence, deployment, development, malicious use, misuse, national security, NIST, risk reduction

March 18, 2025

Patrick J. McGovern Foundation Renews Commitment to Supporting IST’s AI Risk Reduction Efforts

Over the last two years, with the support of the Patrick J. McGovern Foundation, the Institute for Security and Technology (IST) has been on a mission to assess the risks and opportunities associated with the development and deployment of cutting-edge AI foundation models. IST is again excited to announce renewed support from the Patrick J. McGovern Foundation to further advance this vital work.

AI foundation models, artificial intelligence, compliance failure, malicious use, risk reduction

March 18, 2025

Announcing Continued Support from the Swiss Federal Department of Foreign Affairs and the German Federal Foreign Office for the CATALINK Initiative

The Institute for Security and Technology (IST) is pleased to announce the generous political and financial support from the Swiss Federal Department of Foreign Affairs and the German Federal Foreign Office Germany to continue the CATALINK Initiative through 2027.

CATALINK, crisis communications, disarmament, mesh networks, nuclear weapons, risk reduction

January 21, 2025

In the News, Podcast

Strategic Stability: Sahil Shah on Press the Button

Sahil Shah joined Ploughshare Fund’s flagship podcast Press the Button to discuss strategic stability, transatlantic relations, and the need for reliable communications links in all nuclear contexts.

crisis communications, JCPOA, nuclear command, risk reduction

October 11, 2021