DOJ Targets ‘Fake It Till You Make It’ Tech Startups

A top federal prosecutor is reportedly targeting startups that defraud investors ahead of going public.

Ismail Ramsey, U.S. attorney for the Northern District of California, told Reuters in an interview published Tuesday (March 19) that his office’s proximity to Silicon Valley and its investor community makes it uniquely positioned to go after artificial intelligence and other tech startups that dupe backers before their initial public offering (IPO).

“These ‘fake it till you make it’ pre-IPO frauds erode the integrity of public and private financial markets,” said Ramsey, per the report.

He added that startup founders could be tempted to mislead investors about revenue base, product readiness and customer reach to drive up interest in a potential IPO.

Ramsey said in the report that AI will be a focus of his office’s efforts as excitement around the technology continues to build.

“As with any such emerging technologies, AI is fertile ground for fraudsters to make false and exaggerated claims,” said Ramsey, per the report.

The U.S. government has been increasing its scrutiny of the AI sector. Earlier this week, the Securities and Exchange Commission fined two investment advisers for making misleading statements about their firms’ AI capabilities.

The Federal Trade Commission has reportedly seen a jump in AI scam advertisements on social media.

AI has both helped and hindered the global battle against fraud, with the technology giving scammers new tools to trick consumers and businesses.

“That’s because the ability to generate human-like text, virtually clone loved ones’ voices and faces, and scale behavioral-driven attacks increasingly democratized access to cybercrimes that previously only sophisticated bad actors would dream of attempting,” PYMNTS wrote in December.

The technology’s ability to generate code for designing malware has led to a rise in automated phishing attacks and AI-powered business email compromise, as well as automated account takeover attacks.

At the same time, the technology is being used to stamp out fraud, with research by PYMNTS Intelligence showing that rules-based algorithms, AI and machine learning are among the technologies most used to deal with fraud, especially among larger lenders.

“Sixty percent of financial institutions reported using rules-based algorithms to combat fraud, up from 50% in 2022,” PYMNTS wrote earlier this week. “PYMNTS Intelligence also found that at least 66% of financial institutions with more than $5 billion in assets use AI and ML, exceeding the 44% of smaller banks that do the same.”

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.