Korva is a scroll of the internet’s best ideas.
A $250 fee to be a guest
The fake invitation
The invitations praise the author's work and come in the name of the host, the co-host, or one of the producers. An author who writes back is told to pay about $250 or $350 to defray the costs of the interview and support the show. Some pay and lose the money. Fresh Air would never ask a guest to pay, and other NPR shows and podcasts have been impersonated the same way.
Who is behind it
Cybercrime researcher Gary Warner, a founder of the security company Dark Tower, investigated this kind of scam on behalf of authors and found that the phone numbers carried the +234 prefix, Nigeria's country code. In some cases he identified the scammers by name, and they were Nigerian nationals. That points to the Yahoo Boys, a well-known cabal of cybercriminal groups in Nigeria that have been sending email scams for decades, the Nigerian prince emails among them.
AI makes it scale
AI has supercharged it. A chatbot can instantly write a personalized, polished email in perfect English with no grammatical errors that flatters a specific author, and it can do that for hundreds or thousands of targets at once. A Wired writer who gets these emails himself asked ChatGPT to write an email from Terry Gross inviting him onto the show with a $250 production fee, and it did so without hesitation. He then told it the recipient was resisting payment and asked it to persuade him, and it wrote a follow-up arguing that he needed to pay.
Why do scammers charge fees as small as $250?
A small fee feels like an ordinary business expense, so the author is less likely to stop and question it. Across hundreds of targets, even a handful of payments adds up. The scheme is advance-fee fraud, which traces back to the Spanish Prisoner letters of the 1800s, in which a stranger promised a fortune in return for a small payment up front.
Why are Nigerian email scams called 419 scams?
The number comes from Section 419 of Nigeria's criminal code, which covers obtaining property by false pretenses and fraud. The name stuck because advance-fee emails became so closely tied to the country in the 1990s and 2000s. The best-known version, the Nigerian prince email, promises a share of a large fortune in exchange for money paid first.
How does AI make phishing emails harder to spot?
A chatbot removes the spelling and grammar mistakes that once gave scam emails away, and it can tailor each message to a specific person's work. IBM's X-Force researchers reported in 2023 that an AI model needed about five minutes to produce a phishing email that a team of experts spent roughly 16 hours writing. In that test, the human-written email still got a slightly higher click rate.
More from Fresh Air
