Welcome back. Companies will need to find new forms of digital infrastructure and security systems to counter the rise of artificial intelligence-produced deepfakes as more sophisticated methods of breaking into businesses become widespread, according to AI security experts.

“What we need in this moment is new infrastructure,” said Henry Ajder, an AI and deepfake cartographer, speaking at The Wall Street Journal’s CCO Council Summit in London on Wednesday.

“If there was one message I would want people to come away with from today is that your digital infrastructure for ingesting information into your business is almost certainly, woefully out of date with the new age and with the new kinds of capabilities for attackers,” he added.

Read the full story here.

Also today:

• Iran-linked hackers run false-flag operation.
• Pentagon boosts cyber training.
• Healthcare software maker hacked.

The National Institute of Standards and Technology said it will conduct “pre-deployment evaluations” of frontier AI models from Google, Microsoft and xAI to assess cybersecurity risks. To date, NIST's Center for AI Standards and Innovation has completed more than 40 evaluations, "including on state-of-the-art models that remain unreleased," the agency said. (NIST) 

Healthcare software maker Networking Technology is notifying customers of a breach in March that exposed data on some patients, including names, dates of birth, contact information and patient IDs, according to a notification letter reviewed by HIPAA Journal. The company, which does business as RXNT, said it brought in third-party cybersecurity experts to determine the nature and scope of breach. (HIPAA Journal) 

A cybercrime suspect arrested in January by Romanian police was extradited to the U.S. last month over his alleged role in a bank-fraud scheme carried out 17 years ago, the Justice Department said. Gavril Sandu, a 53-year-old Romanian national, is accused of hacking into small-business VoIP systems to obtain login credientials, payment card numbers and PINs, the DOJ said. (Security Week).

The WSJ Pro Cybersecurity team is Deputy Bureau Chief Kim S. Nash and reporters Angus Loten, James Rundle and Catherine Stupp. Follow us on X @WSJCyber. Reach the team by replying to any newsletter you receive or by emailing Kim at kim.nash@wsj.com.