Institute for Security and Technology

Active Defense, the Congressional Agenda, and Teen Hackers: IST at RSA IST will once again join the global community of cyber experts in convening in San Francisco for RSAC this March! From active defense, to the congressional agenda, to stopping teen hackers, the IST te...

#RSAC is less than a month away – is your schedule set? IST will join global #cyber leaders in San Francisco next month, as our team joins sessions tackling active defense, the congressional agenda, stopping teen hackers, and more!
🛡️ Here’s where you can find IST at RSAC:

➡️ Apply now to be considered:

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 de...

The Andrew Carnegie AI–Nuclear Policy Accelerator aims to seed a network of nuclear policy practitioners capable of translating AI developments into concrete policy through technical literacy & applied engagement. Apps for the inaugural cohort close tomorrow, Feb. 28 at midnight PST.
🎓 Learn more:

Imminent National Cyber Strategy May Lean on Offense at the Expense of Defense Reports suggest that the Trump administration's imminent National Cyber Strategy would prioritize cyber offense in order to shape adversary behavior and relegate cyber defense to the back burner. If i...

The Trump admin's National Cyber Strategy is imminent. Will it, as rumors suggest, prioritize #cyber offense over defense to shape adversary behavior? For Claroty #Nexus, IST CSO @megans.bsky.social breaks down the history of cyber strategy & the drawbacks of an offense-first approach.
🛡️ Read more:

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 ...

Moltbook, a Reddit-like forum for AI agents, gained 1.5mil users before research suggested that a multi-agent environment w/o cybersecurity guardrails could be a hotbed for fraud & influence ops. IST’s Gabrielle Tran outlines what this means for the future of cyber-enabled manipulation.
🛡️ Read more:

Episode 251 - Inside the Flock Controversy with Mark Keierleber and Michael Klein

Around the country, Flock cameras have become a source of privacy concerns for parents and teachers alike. IST’s @michaelfklein.bsky.social joined the K-12 Tech Talk Podcast to weigh in on the recent controversy, impacts on schools, and how #securebydefault could be the solution.
🛡️ Listen now:

➡️ Apply by midnight, February 28 to be considered for the inaugural cohort:

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 de...

The Andrew Carnegie AI–Nuclear Policy Accelerator is a initiative designed to strengthen decision-making at the intersection of AI & nuclear policy. Mid-career practitioners across sectors are invited to join this program focused on tech literacy & applied engagement.
🎓 Learn more:

YouTube video by CoSNweb AI and Cybersecurity: What District Leaders Need to Know

IST’s Michael Klein hosted a CoSn webinar earlier this month to brief district leaders on the state of AI & #cybersecurity in ed. From AI development history to how threat actors are using it to their advantage, he walked leaders through tackling AI concerns in their districts.
🛡️ Watch the webinar:

“Across these shifts, it becomes increasingly clear that U.S. compute-related export controls are becoming less about static rule-writing and more about sustained operational oversight,” they write.

IST authors Jennifer Tang and Gabrielle Tran draw from IST working group discussions to present stakeholder reactions, policy implications, and key takeaways from the last two months in AI export control.

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...

🚨 NEW from IST: Jan 2026 was, in many ways, an inflection point. From the resumption of Nvidia H200 sales to China to congressional discussion on restricting remote cloud access, a new memo from IST’s #AIChipExportControl Initiative unpacks what these developments mean for the future.
🏛️ Read more:

➡️ Apply by midnight, February 28 to be considered for the inaugural cohort:

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 de...

As AI tech enters nuclear weapons, policy debates are accelerating faster than practitioners can keep up. The new Andrew Carnegie AI–Nuclear Policy Accelerator at IST seeks to close this gap, providing mid-career professionals with hands-on technical literacy & applied engagement.
🎓 Learn more:

UnDisruptable27 UnDisruptable27 (U27) exists to translate high-level risk awareness into local readiness. Focusing at the nexus of water infrastructure and emergency healthcare, this initiative responds to an urgent ...

“Even when compared to other critical infrastructure (CI) sectors, water stands out as particularly “target rich and cyber poor,ˮ’ the report reads, citing @joshcorman.bsky.social and #U27.
🛡️ Learn more:

In the Almanack, DEF CON Franklin called out IST’s #UnDisruptable27 Initiative, alongside fellow #CyberCivilDefense grantees, including the @cyberpeaceinst.bsky.social and @cltcberkeley.bsky.social Cyber Resilience Corps.

At @defcon.bsky.social last summer, hackers of all kinds convened to share #cyber research. The Hackers’ Almanack translates this work into plain language to give policymakers a window into the latest in offensive AI, cybercrime, critical infrastructure vulnerabilities & more.
➡️ Read the Almanack:

Cybersecurity Community Celebrates Winners of Third Annual Cyber Policy Awards Members of the cyber policy community were honored at the Third Annual Cyber Policy Awards on Thursday, February 5, 2026 before a packed house of cyber community veterans, leaders, and industry partne...

🏆 Meet the winners of the Third Annual Cyber Policy Awards:

At the Third Annual #CyberPolicyAwards, NJ CISO Michael Geraghty was awarded the Cyber Policy Award for U.S. Domestic Policy Impact. NJOHSP director Thomas Hauck congratulated Michael, commending his “exceptional leadership & vision in advancing New Jersey’s cybersecurity capabilities.”
➡️ Read more:

🎓 Apply by midnight, February 28 to be considered: docs.google.com/forms/d/e/1F...

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 de...

The Andrew Carnegie AI–Nuclear Policy Accelerator is a new, practitioner-focused initiative designed to strengthen decision-making at the intersection of #AI & #nuclear. Are you a mid-career professional in the national security space seeking hands-on technical literacy & engagement?
🎓 Learn more:

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 a...

The future with AI in it is filled with “known unknowns” and “unknown unknowns.” For the #NatSpecs blog, IST’s Mariami Tkeshelashvili reflects on her co-authored report, released today, and what she sees as our #AI responsibility.
❇️ Read more:

Also in this edition:
+Dr. Natalie Sullivan joins @brysonbort.bsky.social on @hacktheplant.bsky.social to discuss cascading failures in our public health systems
+Nick Leiserson & Taylor Grossman urge policymakers to “wake up” to the Res. Proxy threat
+Winners of the Third Annual #CyberPolicyAwards!

An AI Loss of Control “Warning Shot” How might you lose control of a system as complex as a frontier AI model? To quote Ernest Hemingway’s The Sun Also Rises, the same way you go bankrupt: “Gradually, and then suddenly.” Contrary to what...

Welcome back to The TechnologIST! This month, we’re looking at 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.
🗞️ Read more:

Thank you to @pjmfdn.bsky.social for their generous support of IST’s continued AI research!

“As AI capabilities advance, monitoring AI LOC risk is essential to maintaining situational awareness. Only with collective efforts can we ensure that humanity harnesses AI capabilities in ways that lead to flourishing, rather than harmful outcomes,” the authors write.

The Indications & Warning framework—drawn from the Intelligence Community— identifies 7 potential indicators of AI LOC risk, including scheming, manipulation, and self-preserving behavior, and establishes 5 severity warning levels for risk.

In “AI Loss of Control Risk: Indications & Warning,” authors Mariami Tkeshelashvili, Ritika Verma & Steven M. Kelly present a framework to enable orgs to assess the current risk landscape, implement proportionate safeguards, and align stakeholders on response before critical thresholds are crossed.

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 stakeholde...

🚨NEW from IST: Though tech & policymakers alike are eager to address AI Loss of Control, in which an AI system diverges from authorized constraints, gaps still exist in how stakeholders understand & perceive these risks. Today’s report proposes a new methodology to monitor AI LOC risk.
❇️ Read more:

Why AI Is a Big Problem for School Cybersecurity Many school districts are ill-prepared to defend themselves against AI-powered cyberattacks.

As AI techs proliferates, so too do #cyber attacks, including those against schools, IST Senior Director @michaelfklein.bsky.social told EdWeek. AI tools greatly enhance the capabilities of the average hacker, lowering “the barrier for sophisticated cybercrime,” he explained.
🛡️ Read more: