Work in AI? You are at risk of being honey-trapped
What's the story
Aaron Brown, a former senior CIA officer and founder of software company Lumbra, has warned that people in the AI industry are at risk of honey trap schemes. He said these tactics have traditionally targeted politicians and military personnel but are now being used against those in charge of valuable corporate secrets. Brown said spies target CEOs because they can access sensitive information like roadmaps, export decisions, and customer lists.
Scandal
NVIDIA CEO embroiled in scandal
Brown's warning comes in the wake of NVIDIA's Jensen Huang being embroiled in a honey trap scandal.
He has been accused of having an affair with Alice Huang, a former executive at Singapore neocloud who is suspected to be a Chinese spy.
She is also accused of smuggling billions worth of restricted NVIDIA chips into China.
Precautionary measures
OpenAI warning its employees about potential honey traps
Following the Huang scandal, OpenAI has started warning its employees about potential honey traps.
The company is cautioning them against outsiders trying to befriend them for confidential information.
This highlights that anyone with access to sensitive data, from board members to executive assistants, could be a target of such schemes.
Deceptive tactics
How to identify a honey trap?
Brown revealed that honey traps disguise themselves as investors, customers, recruiters, or consultancies.
Their requests seem legitimate on paper and don't require their targets to fall in love.
The goal is to make the target feel comfortable talking with them.
This tactic is more effective because people aren't looking for hidden agendas in these seemingly professional interactions.
Tech exploitation
Irony of tech founders being vulnerable to oldest scam
Brown noted the irony of tech founders digitizing human life while being vulnerable to the oldest scam in the book.
He said, "Why fight through firewall when you can get close to the person who can open doors, influence decisions, or make requests nobody wants to question?"
The rise of AI has made honey trapping easier by enabling quick target research, fake social profiles creation, and even fake voice calls, said Perry Carpenter, chief deception strategist at KnowBe4, to TechCrunch.
Warning signs
Red flags to watch out for
Once targets realize they've been compromised, the threat escalates as attackers have a means of blackmail.
Carpenter advised looking for unusual patterns such as "coincidental encounters that happen a little too often."
Other red flags include persistent curiosity about work and colleagues, secrecy about meetings, and pushing for introductions or information.
Brown added if someone's requests are getting bigger than usual, it could be a sign of honey trapping.