AI warning: Federal officials warn of growing threat from artificial intelligence

CHICAGO (WLS) -- Former Homeland Security officials and federal cybersecurity agencies are warning that artificial intelligence is increasingly being exploited by terrorists, criminals and foreign adversaries, raising concerns about potential threats to national security.

Former Homeland Security Acting Undersecretary for Intelligence and ABC News contributor John Cohen told the ABC7 I-Team that terrorist groups and criminal organizations are already using artificial intelligence to enhance their capabilities.

"Terrorist groups and criminal organizations are already using artificial intelligence to improve their ability to conduct cyber attacks, to develop detailed attack playbooks, to better understand how to construct weapons that can be used," Cohen said.

The growing use of AI presents significant national security concerns because it can accelerate the planning and execution of attacks, according to Cohen.

"Artificial intelligence has dramatically shortened the time it takes for a terrorist group to plan an attack, and it increases the likelihood that their attack plan will be successful," he said.

Cohen said the technology has raised concerns about potentially devastating and unpredictable attack scenarios.

"The attacks of 9/11 took years to plan, prepare for, and ultimately implement. Today, an unsophisticated, angry, socially isolated individual... can develop a detailed attack plan, including instructions on how to construct an explosive device or some other lethal weapon, in a matter of minutes," Cohen said.

SEE ALSO | OpenAI flags concerning new AI behavior and vows to track it more closely

Federal authorities, including the National Security Agency, have warned that adversaries such as China are working to exploit American-made AI technology and extract its capabilities.

AI company Anthropic recently announced it had thwarted numerous misuse cases involving its Claude AI system. According to the company, those attempts included efforts to reverse engineer offensive drone technology, surveil government officials, automate disinformation campaigns and scam people out of money.

"Part the challenge facing law enforcement as artificial intelligence becomes more sophisticated and more ingrained in our society is to understand how the criminals and threat actors are using it, and then to use the technology to counteract those illegal or threat-related activities," Cohen said.

The warnings come as concerns continue to rise after cyberattacks targeting critical U.S. infrastructure, including attacks linked to Iranian groups, raising questions about the security threats of emerging AI technologies.