At first, I was worried. But something didn't seem right. She is a friendly and well-connected person. If she were stranded in London, then why ask for help on Facebook rather than call the people she visited there? What about the local embassy?
I continued to chat; she said her credit card, cash, and cell phone were all stolen. I asked if she still had her passport, which she did. Her issue was having to check out of her hotel and not being able to pay. She added a time constraint of having to fly out in two hours, with the promise of paying me back.
HOLD ON! What just happened here? How did I commit to send money? Did I?
She continued to ask if I could help. I wanted to -- I really did. But asking me for money? That is always an instant red flag. She eased her pushing when I told her I would be able to help the next morning, though would be able to spare only $100 (USD), not the $500 she needed. She said she would have to postpone her flight, and then very helpfully asked me for my ZIP code so she could help me find a local Western Union branch. When I didn't give her my ZIP code, she gave me the Web site to search for it myself. Right.
I offered to try and connect her with other mutual friends -- fellow organizers of science fiction conventions (SMOFs) -- who might be able to help her. She sidestepped that suggestion by saying, "I don't want to worry anyone."
I convinced her the situation was beyond worrying, and asked how the SMOFs could contact her. She repeated that her cell phone was stolen, so I asked for her hotel's number and room. Instead, she kept going back to her need for the money -- saying she was "shaken" and "recovering." I kept going back to the need for contact information. She eventually gave me an address in London to wire the money to, though when I again asked for her hotel contact information, she offered an email with a free service.
I continued to question her until she asked if I was trying to verify she was who she said she was. When I replied in the affirmative, she removed me as a friend on Facebook, which prevented me from writing on her wall. I guess some people might run right back at this rejection, but I didn't.
Still, I was worried, so I emailed my fellow SMOFs. Sure enough, she was clearly not who she said she was. Her account had been taken over, and I spent an hour -- late for a date -- chatting with a scammer.
How do I know for sure?
1. Her English is better than this. 2. She is far from an helpless person 3. She has local friends. 4. She refused to identify herself or help me do so. 5. The name she responds to is not the one I used. 6. When contacted on the phone by a fellow SMOF, she verified her account had been hacked.
I IMed a friend at Facebook security and asked if there were any ongoing 419 scams. His immediate reply: "Held at gunpoint in London?"
The person I spoke with may not be a person at all, but a chat bot. I can not prove this conclusively in either direction.
Scams and chat bots are not new criminal technology. Using new media to perform these attacks catches people unprepared. What's interesting is that if this really was a bot, it passed the Turing test well enough to leave me worried about my friend an hour later.
Be careful out there, folks. Don't ever send money online if you can avoid it, and verify people are who they say they are. Facebook has a security page covering these issues here.
Follow Gadi Evron on Twitter: http://twitter.com/gadievron
Gadi Evron is an independent security strategist based in Israel. Special to Dark Reading.