Automated Scammer Responses

I read this OP, about wasting an email scammer’s time by responding to them and playing through the scenario. And it made me think of this RAND report about how to fight Russian propaganda, in essence, to match kind with kind.

Scammers rely on the scale to be profitable. They need to be able to flood the internet with enough scam emails in order to find the 0.1% stupid enough to fall for it and actually go through with the scheme.

Right now, this is very easy for them. They send out 10,000 emails and they get back 5 responses. 4 of those responses will be real and 1 will be someone wasting the scammer’s time. The scammer’s time involvement is sufficiently low - writing an email every few days - that this doesn’t really register. I doubt they even care.

But what happens if, in response to their 10,000 emails that they send out, they get back 10,000 answers, each of which looks like a real response from a real human, and the scammer has no way of knowing which one is a real person? He’ll know that 4 are real, but he won’t know which ones.

The scammer has to device a response that he can blast out to all 10,000 of his responders. He can’t hand-write each one. He has to maintain a conversation with the mystery four people, without being able to identify them, and that’s a lot harder. The four will try to ask questions, try to hold a conversation, and they’ll just get back impersonal demands to send money. The number of people that the scam will successfully work on drops from 0.1% of the population to 0.01% of the population, because scamming someone requires building up a relationship and you can’t do that via cut-and-paste responses.

Resolved: The Federal government should develop a program to develop AI / chatbot technology to detect and respond to scammers in bulk.

How much is this going to cost the Federal government?

Why does the federal gvt need to do this? Seems this could well be a ‘service’ or feature built into mail clients/service.

Secondly - think about the overall affect on network traffic if suddenly you get into bot wars sending and recieving spam to fight spam.

This sounds like a job for the email programs, not the government - you’d want this to occur in the same place spam filtering occurs. And it doesn’t have to be all that good at replicating a human; it just needs to be good enough to pass the spammer’s spam filter, and clog his inbox. All the titles would just be “RE: The spammer’s title”, so he’d have to open each one individually to see if it’s real…

  1. This sounds like a job for patreon or kickstarter.

  2. …or New Zealand: New Zealand’s Netsafe Organization has devised a simple and clever way to respond to these scam messages. It is called REscam, and all you need to do to participate is to forward any of the scam messages to, and the bot gets to work, engaging in a phony conversation with the scammers with an endless stream of “questions and anecdotes” so the scammers have less time to pursue other people.

  3. The go-to location for scambaiting is 419eater dot com.

Seems like an awful lot of effort to protect .1% of the population “dumb enough” to fall for it.

  1. Most crimes only affect a small subset of the population. We could decriminalize murder, for example, and just let people handle revenge themselves.
  2. The money raised from these sorts of ventures are fairly likely to end up going towards promoting other crimes. It’s all just revenue streams for mafia and mafia states, who then use that money to corrupt politicians, assassinate clean politicians, and bend the laws in the aim of making crime easier and more profitable. If you can, fairly cheaply, cut off these revenue streams, then you’re spared having to figure out how to deal with a criminal organization that’s established in a foreign nation state.
  1. It involves billions of dollars.
  2. The vics are disproportionately elderly; I suspect many are on the road to dementia and have organically reduced defenses.
  3. Some of the scams are better than others. Spear phishing ensnares small businesses for example.

Sage: What we really need is for the FBI and FTC to increase their white collar crime units and attack robocalling and cyberscams. They could work and in glove with the scam baiters, who would welcome them with open arms. I suppose the NSC could provide a few R&D grants for bot development, but I don’t see that as a big issue. Baiter bots exist already, and nonprofit resources would flow to them if the Feds would raise their profile on this issue.

After 911 the FBI’s white collar crime division shrank as they ramped up anti-terrorism investigations. Fair enough. But I say adding a couple of hundred investigators to their white collar crime units would be a good step. We have a problem with financial crimes as well.