Episodios

  • Sara Hooker on compute thresholds for AI; CrowdStrike breaks the internet

    Sara Hooker on compute thresholds for AI; CrowdStrike breaks the internet
    Jul 25 2024
    AI governance regimes the world over have seized on compute thresholds as a mechanism for implementing various controls on artificial intelligence systems. The basic idea is that if an AI model relies on a sufficiently large amount of computing power, then various controls kick in. As AI models get larger, the thinking goes, they also get riskier, and this means that AI governance regimes should focus on the largest models, as measured by computing power. But does this idea make any sense as a governing tool for the models of today and tomorrow? Sara Hooker leads Cohere’s research operation, and she’s looked hard at whether compute thresholds can be applied to AI systems to mitigate risks. On this episode of Safe Mode, she sits down with host Elias Groll to discuss her research on compute thresholds. CyberScoop’s newly minted editor in chief, Greg Otto, also joins the show to discuss how an errant CrowdStrike software update took down a huge number of critical services across the internet. Links: On the Limits of Compute Thresholds as a Governance Strategy | arXiv CrowdStrike Falcon flaw sends Windows computers into chaos worldwide | CyberScoop
    Más Menos
    46 m
  • Dewey Murdick on enabling principles for AI governance; a landmark breach at AT&T

    Dewey Murdick on enabling principles for AI governance; a landmark breach at AT&T
    Jul 18 2024
    Around the world policymakers seem to be grappling with the same question: How best to govern artificial intelligence? The Center for Security and Emerging Technology, a Washington-based think tank, has emerged as a key and influential player in that conversation. And on this episode of Safe Mode, its executive director, Dewey Murdick, sits down with host Elias Groll to describe what he sees as the enabling principles of AI governance. Also on the show this week, CyberScoop reporter AJ Vicens discusses a landmark breach affecting AT&T that saw call and text records of nearly all the company’s customers be stolen.
    Más Menos
    40 m
  • Allan Liska on why it’s time to ban ransomware payments; Russian info ops

    Allan Liska on why it’s time to ban ransomware payments; Russian info ops
    Jul 11 2024
    We are living through an epidemic of ransomware. Last year alone, the amount paid to ransomware operators exceeded $1 billion, and the entities getting hit and seeing their operations disrupted by ransomware included hospitals, schools and libraries. Ransomware is both a public-health and national-security crisis, yet efforts to address the problem are not making a dent. Allan Liska is a senior intelligence analyst at Recorded Future and a leading analyst of the ransomware phenomenon, and on this episode of Safe Mode he sits down with host Elias Groll to discuss why it’s long past time to ban ransomware payments. CyberScoop reporter Derek Johnson also joins the show to discuss the U.S. intelligence community’s observations of information operations targeting the 2024 election.
    Más Menos
    38 m
  • Selena Larson on e-crime matching nation state hackers; Disinfo before the Supreme Court

    Selena Larson on e-crime matching nation state hackers; Disinfo before the Supreme Court
    Jun 27 2024
    When studying cyberoperations, it’s easy to become enamored with state-backed hackers. Russian and Chinese operations have done much to shape our understanding of how power is exercised online, but it’s not clear that the intense attention dedicated to such operations is well-placed. Amid the current epidemic of ransomware, criminal hacking groups are arguably having as much if not a greater impact on the lives of ordinary people. On this week’s episode of Safe Mode, Selena Larson, a senior threat intelligence analyst at Proofpoint, sits down with host Elias Groll to make the case for why e-crime groups should be afforded just as much attention as state-backed hackers, reframing how we think about the cybersecurity landscape. CyberScoop reporter Derek Johnson also joins the show to discuss the fallout of a major Supreme Court ruling on the fight against disinformation.
    Más Menos
    44 m
  • Protecting teens against sextortion; threats to election workers

    Protecting teens against sextortion; threats to election workers
    Jun 20 2024
    The proliferation of artificial intelligence is exposing teens to a new online threat: AI-generated sexual imagery. Heather Barnhart is a fellow at the SANS Institute and a digital forensic expert, and on this episode of Safe Mode she sits down with host Elias Groll to discuss how teenagers and their families can protect themselves against the scourge of sextortion. CyberScoop reporter Derek Johnson also joins the show to discuss how election administrators are leaving their posts due to widespread harassment.
    Más Menos
    35 m
  • Keeping Ukraine’s grid up and running amid war; Snowflake customers under attack

    Keeping Ukraine’s grid up and running amid war; Snowflake customers under attack
    Jun 13 2024
    Following Russia’s full-scale invasion of Ukraine, the Ukrainian power grid has come under sustained attack by Russian forces, and keeping it up and running has been a significant challenge. One of the major issues Ukrainian energy officials have had to confront is the way in which Russian electronic warfare systems have disrupted GPS, which is a key tool in running electrical distribution systems. Taras Vasyliv, the head of the Dispatching Control Systems Department at the Ukrainian energy utility Ukrenergo, and Joe Marshall, a security researcher at Cisco, collaborated on an innovative solution to address the vulnerabilities of GPS systems in the Ukrainian grid. And on this episode of Safe Mode, they sit down with host Elias Groll to discuss the novel tool they developed to maintain the delivery of electricity even in the face of Russian jamming. CyberScoop reporter AJ Vicens also joins the show to discuss a wide-ranging operation targeting customers of the data storage company Snowflake. As many as 165 companies ‘potentially exposed’ in Snowflake-related attacks, Mandiant says | CyberScoop
    Más Menos
    31 m
  • Josh Harguess on AI red-teaming; an Israeli info op targets the U.S.

    Josh Harguess on AI red-teaming; an Israeli info op targets the U.S.
    Jun 6 2024
    To address AI risks, policymakers and technologists around the world have embraced the concept of AI red-teaming — the adversarial testing of AI models. But AI red-teaming is such a new concept that few people know how to conduct such tests. The discipline of AI red-teaming very much remains under development, but as AI systems are rolled out far and wide, this discipline is only going to become more important to make sure that AI tools work as intended. On this episode of Safe Mode, host Elias Groll sits down with Josh Harguess, the AI security chief at Cranium AI who is one of the few bona fide experts in the field, to dive deep on how to carry out AI red-teaming. CyberScoop reporter Derek Johnson also joins the show to discuss an Israeli information operation targeting audiences in the United States and Canada.
    Más Menos
    37 m
  • Joe Sullivan on his legal battle with the feds; the latest on Russian influence operations

    Joe Sullivan on his legal battle with the feds; the latest on Russian influence operations
    Jun 3 2024
    Joe Sullivan on his legal battle with the feds; the latest on Russian influence operations by Safe Mode Podcast
    Más Menos
    47 m