Top U.S. official warns that computer systems across the country need to strengthen cyber defenses to counter AI risks The U.S. government's top information security official has warned that computer systems across the country need to strengthen their defenses to counter the growing threat of artificial intelligence (AI)-driven cyber attacks.

Senior U.S. official warns that computer systems across the country need to strengthen cyber defenses to deal with AI risks The U.S. government's top information security official has warned that computer systems across the country need to strengthen their defenses to deal with the growing threat of artificial intelligence (AI)-driven cyber attacks. Speaking on Wednesday, acting U.S. federal chief information security officer Michael Duffy said government and private sector agencies have typically taken a reactive approach to cybersecurity in the past. He said that in the AI ​​era, this approach must change in order to take preventive measures before hacker attacks occur. "Given the AI ​​capabilities we're seeing now, we probably won't have time to clean up the mess," Duffy said during a panel discussion at a cybersecurity conference in Las Vegas. "Policies for the next decade can't always be passive after a major incident happens." Rapid breakthroughs in AI systems, including Anthropic PBC's Mythos model, have raised concerns that hackers could use the technology to launch destructive cyberattacks on critical infrastructure. The Trump administration has committed to taking action to address the cyber risks posed by AI and recently completed a voluntary security testing framework for cutting-edge models. Duffy's remarks came shortly after Bloomberg News reported that hackers had launched a series of sophisticated attacks on a number of Wall Street companies in recent days, highlighting the increasingly sophisticated cyberattacks and fraud threats targeting financial services companies. In the same panel discussion, another senior U.S. Department of Homeland Security official emphasized that the United States must not only maintain its leading position in the global AI field, but also ensure that the opportunity to use this technology is not only in the hands of a few. Joseph Alm, assistant secretary for cyber, infrastructure risk and resilience at the Department of Homeland Security, said that with the rapid development and popularization of AI, the industry needs to prepare for a period of greater vulnerability to attacks. "The loss of network systems is no longer a black swan event, but will become the norm," Alm said. "You need to change your thinking and realize that in an era when everyone can discover or exploit vulnerabilities, everyone will face a period of vulnerability that lasts for several years." Alm added that after a few years, the situation may gradually return to a more normal state. via cnBeta.COM - Chinese industry information website (author: Source: Global Market Report)