World Trade Center Building 1 is located on the site of the 9/11 attacks
Phoenix Technology News: On September 1, Beijing time, according to CNBC, the U.S. House of Representatives‘ Permanent Special Intelligence Committee urged U.S. intelligence agencies to better guard against “black swan” AI risks. The committee warned in a new report that terrorists or hostile forces could use this technology to develop more dangerous weapons and plan more deadly attacks.
The report said: “The committee believes that the intelligence community and broader policymakers must carefully consider how to prevent AI tools from being exploited by malicious actors, including terrorists.” The report aimed to assess the measures taken by the United States over the past 25 years to respond to intelligence errors that ultimately led to the September 11, 2001, terrorist attacks.
“Since 9/11, preventing terrorist organizations from acquiring the equipment and expertise needed to manufacture and use weapons of mass destruction has been an important focus of work in the intelligence community and national security agencies. The potential of cutting-edge large-scale language models could make it easier for any malicious actor, including terrorists, to plan and carry out more destructive attacks,” the report said.
This report and its warnings about the disruptive potential of AI are less than two weeks away from the 25th anniversary of the 9/11 terrorist attacks that killed thousands of people and reshaped the U.S. security system. Meanwhile, AI is rapidly developing and driving massive economic activity, which has forced lawmakers and officials to deal with thorny challenges surrounding regulation of this technology.
The report was co-authored by Rick Crawford, a Republican Party of Arkansas congressman and chairman of the House Intelligence Committee, and senior members of the committee, Jim Himes, a Connecticut Democrat, Elise Stefanik, a New York Republican, and Josh Gottheimer, a New Jersey Democrat. They warned that existing security measures might not be sufficient to keep up with the rapid pace of AI development.
“Although AI labs are trying to prevent models from providing information that might help illegal actors implement related behaviors, such as creating biological weapons, the underlying capabilities of the models are developing rapidly, and it is very difficult to fully control these uses of the models,” these legislators wrote.
These legislators urged the U.S. intelligence community to “accelerate responsible adoption of advanced AI capabilities to ensure that the U.S. remains ahead of its rivals.” The report also noted that investing in “security AI tools for information collection, analysis, and early warning, combined with rigorous testing, human supervision, and robust privacy and civil liberties protections” is critical to responding to the next generation of threats.
The committee overall believes that the United States “is facing the most complex and dangerous security threat environment since September 11, 2001.” The report states: “We still have a serious gap in how to respond to and address these weak links. We are not fully prepared to respond to and address our own vulnerabilities.”