Q&A With Jason Matheny, Founding Director of the Center for Security and Emerging Technology - Georgetown University
webA 2019 Q&A with CSET founding director Jason Matheny explaining the center's mission to bridge AI policy and national security research, relevant to AI governance and safety policy infrastructure.
Metadata
Importance: 38/100interviewnews
Summary
Jason Matheny introduces CSET's mission to provide rigorous AI policy analysis for national security policymakers. He highlights that current AI systems are brittle and fragile rather than superintelligent threats, and emphasizes the need to understand AI failure modes. CSET aims to fill a gap between rapidly growing policymaker demand for AI analysis and the think tank community's limited supply.
Key Points
- •CSET focuses initially on AI policy because demand for analysis has outpaced supply in the think tank community over the prior four years.
- •Key national security concerns include U.S. competitive position in AI capabilities, workforce, data, and hardware relative to other nations.
- •Matheny argues current AI systems are brittle and fragile, not powerful Skynet-like threats — more likely to cause failures than existential catastrophe.
- •Priority should be understanding AI failure modes and building more robust, secure systems while cooperating internationally on safety and ethics.
- •CSET was described as the largest U.S. center focused on AI and policy at its founding in 2019.
Cached Content Preview
HTTP 200Fetched Jun 15, 202612 KB
Category: Discovery & Impact
Title: Q&A With Jason Matheny, Founding Director of the Center for Security and Emerging Technology
Date Published: February 28, 2019
Share
Share this on Facebook
X
Share this on X
Share this by Email
The founding director of the new Center for Security and Emerging Technology, answers a series of questions.
Jason Matheny, founding director of the new Center for Security and Emerging Technology in the School of Foreign Service, explains the unique role Georgetown has to play in bridging the gulf between technology and policy and providing critical insight and analysis to inform the development of Artificial Intelligence (AI) policy.
Q: What is emerging technology? Where will CSET focus first?
“Emerging technologies” is a general term used to refer to rapidly developing technologies such as artificial intelligence (AI), biotechnology and nanotechnology. AI is the use of machines to accomplish cognitive tasks such as perception, reasoning, problem solving and planning, and it’s a general purpose technology with many applications ranging from medicine to transportation to manufacturing, law and warfare. CSET decided to focus on AI for our first two years for two reasons. First, AI is an important topic that will have broad effects on security. Second, AI is a topic where the demand for policy analysis has grown much faster than the supply –in the last four years, policymakers have been increasingly interested in how to approach AI and policy, but the think tank community has had trouble keeping up. Before CSET spreads out to other topics, we wanted to ensure we’re keeping pace with the needs of policymakers related to AI.
Q: What challenges and opportunities does AI present and what are policymakers struggling with?
There are a range of challenges related to AI, but national security is a critical area of focus. Where does the U.S. sit competitively relative to other countries in its capabilities, in its workforce, in the amount of data that we have that we can use to train AI, and the amount of hardware and computers that we can use in order to run AI? There are specific application areas that are relevant to national security, such as cybersecurity, intelligence and systems for analysis and collection, as well as AI that is embedded in weapon systems of competing nations, that we have to be aware of, and able to develop countermeasures to.
Q: AI is in the news quite a bit lately. How accurately is it presented in the media, especially in terms of capabilities and threats?
There’s a lot of hype right now about AI in the popular press. Sometimes the capabilities of current systems are exaggerated — the things that we should worry the most about related to AI is not how powerful these syst
... (truncated, 12 KB total)Resource ID:
fde0760f14ccb4b1 | Stable ID: sid_KARH88Ngog