Social media is flooded with fake accounts. One way of solving that riddle is to introduce a scale, or a score, that will verify that the account is a real person or business.
I have occasionally posted articles on fake accounts. I call them “phantoms”. My most recent encounters involved Facebook’s Marketplace. I would safely estimate that 50% of my inquiries came from fake accounts. Facebook acknowledges that this is a major problem, frequently purging millions of fake accounts. All social media platforms have this problem.
A recent news event provides some insight on how this is affecting businesses. Doordash came into its own during the pandemic as millions of people solved the problem of feeding themselves by calling restaurants that delivered their food to their customers. Doordash made this easier for both customers and businesses. This produced three concerns. “Real people” created certain risks for delivery people, and vice versa. Secondly, fake customers would emerge as usual, often pumping up reviews or generating false narratives about a business. Finally, fake businesses would emerge. The latter was a real problem because it dealt with real money, real fraud and considerable losses. The case of a high-end restaurant in Anchorage, Alaska is one example. They don’t deliver meals. Yet someone spoofed their name, borrowed photos and menus from their website, and took people’s money. It took persistence on the part of the owner to get the account removed from Doordash and it created considerable confusion for some of the restaurant’s customers.
I don’t do Doordash, so I will stick with Facebook for this discussion. But what both have in common is the desire to minimize, if not eliminate, phantoms. How can this be done? I have one suggestion that would help – scoring.
A Verification Scale
Scoring is basically a verification scale. For example, on a scale of one to ten, one is “low authenticity” while ten is the highest level of authenticity possible. Using Facebook as an example, a user with a score of one is most likely someone who has just created an account, has no friends, has only one or two profile pictures, no photos posted and no history of posting activity. Facebook, using their sophisticated algorithms, could also use negative scoring such as the discovery that profile pictures are duplicated or possible duplicate accounts existing for the user in question. Conversely, a person with a score of eight may be someone who has a long history of participation on Facebook, numerous friends and actively follows various groups and people. Another element may be the degree of financial activity they have with Facebook such as paying for promotions.
Twitter (aka X) and Gab (a competing chat platform) have addressed this issue by introducing paid subscriptions, denoted by a blue bird or green checkmark respectively. This indicates to readers that this account is verified at least to the degree there is someone behind the account that is willing to pay a small fee. But as Twitter discovered, fraudsters found the $84 annual fee at Twitter incidental. A fee should be only part of the authentication scale. Twitter and others have algorithms running that routinely identify what is most likely “bot” activity, comments and such that are generated by computers, not by people. But it is clear that social media platforms still struggle with truly authenticating users. This is especially problematic on platforms where users utilize pseudonyms to disguise their identity. There are cases where pseudonyms are useful. Telegram, for example, is one the more popular chat platforms on the planet, particularly in Russia. Would you put your name on a posting in Russia? So how a scoring scale is engineered would differ from platform to platform.
Scoring would address this problem, not solve it. It would provide readers with an indicator of whether the posting is legitimate. For Facebook Marketplace, it would immediately provide a reliability indicator. A score could even be based on dynamic factors. For Marketplace, one reliability indicator is whether they are local. I could check a box “local only” on an item I am selling and this could be factored into the score when a prospective buyer makes an offer. So when a scammer from Russia wants to buy my electric lawn mower, the reliability index would be lowered or the account even blocked. Most phantom accounts would have low scores because they are recently created, have no followers, no posting activity nor personal info. The reliability score would most likely be a “one” or a “two”, and I can choose to ignore the inquiry without having to investigate the user’s profile. That would save me a lot of time and certainly waste the time of scammers. Another idea is to place a floor on the minimal reliability score you will accept, such as “three” or “four”. This would virtually eliminate all phantom buyers.
Scoring is actually used on some sites I have used in the past. StackExchange is one that is heavily weighted toward active participants. Serious participants can advance their score by being engaged with the platform. They gain points by how they engage others and whether their solutions are found to be testable and repeatable.
As you can surmise, scoring is not used to censor or exclude anyone. Every “real” person or business can enhance their score by using the platform as intended. Each platform can determine their own standards. Using the Doordash example, the highest level of verification would be some independent verification from business owners. Scoring would obviously indicate the participation level of the user. Ultimately, it leaves it up to the user whether they want to trust this person or not.
At the highest level, individuals can subscribe to their own digital certificates. But anyone who has gone through this process can tell you that it is costly and has its own risks. The idea has been kicked around in Europe to require a digital certificate to gain access to the Internet which brings up visions of Revelations 13 (666). Yet a digital certificate can be used as a positive for scoring. Whether it is Facebook, Twitter or Doordash, a digital certificate would be helpful. As best as I can tell, none of them incorporate digital certificates.
The Danger of Scoring
Anyone who is following this discussion probably has the same concern as I. Scoring can easily morph into “social credit” scores. This is done with Orwellian efficiency in China and it is used in some degree by various corporations and credit agencies. Paypal and Bank of America have both used social credit scoring to cancel or suspend customer activity. If scoring is to be acceptable, it must avoid applying social and political criteria. The purpose of scoring should be simple — to verify whether the person you are corresponding with is real. That’s all. Adding “community standards” or “fact-checking scores” to individuals complicates things. Someone like John Stossel, for example, is “verified” in Facebook yet he has also been shadowbanned and “fact-checked”. What people need to know is simply whether this John Stossel is the John Stossel. Nothing more.
Since I am focusing on Facebook, I decided to explore whether Facebook offered any high-level verification tools. Turns out, they have one. So I gave it a shot.
There is a Verification Form:
- Requires an identification such as driver’s license, recent utility bill, passport
- Suggested you have alternative sites such as LinkedIn, Substack or your own website
- Tone of the form appears to cater towards “celebrity,” but I am testing it out as someone who is a public presence, while not well known.
The answer was immediate:
Thanks for submitting your request for verification on Facebook. Our team carefully reviewed your account and determined that unfortunately it’s not eligible for verification at this time.
We understand that this can be frustrating, but we nonetheless encourage you to continue growing your public presence. You can submit another verification request after 30 days.
Please feel free to review our requirements for verification at the Help Center:
We want to protect your authentic presence. If you think you’re being impersonated, please let us know by visiting the Help Center:
As is often the case with social media, it is impossible to gather an explanation from this response, which took less than five minutes. And I have to wait another 30 days? Sounds like a lifetime of futility.
In summary, it appears that this “verification” option is not readily available to everyone. But I will keep an eye out for folks to see if this “verification” check appears on other people’s accounts.
LinkedIn has a verification using CLEAR. Not clear on exactly how this is done (instructions were not consistent with what was being presented on-screen).
Gab indicates I have been “verified.” I paid for an advanced account at one time, but then stopped. The green checkmark remained.
Substack’s is self-generated based on paid subscribers (which means a lot of subscribers).
It appears that scoring is about the only way to painlessly and universally provide a measure of a person’s authenticity. The ways that social media platforms “verify” someone is mostly through a subscription fee or restricted to celebrities seeking brand-name protection. Digital certificates are nowhere to be found on social media sites.
About the best option for the average user is to develop their own scoring system. I would recommend you do this for each platform you use. I provided some hints above, but here are a couple more you can add and the list below can be augmented by your own criteria.
- Age of account
- Followers or friends
- Most recent posting
- Clarity of profile
- Cross-platform presence (on other social media platforms)
- Their own website
- For high risk transactions (like buying or selling a car), consider third-party verification resources such as Better Business Bureau, local Chamber of Commerce, and public records.
Finally, it must be stressed that scoring will not prevent fraud, but it is a valuable tool in reducing the risk of corresponding with fraudulent (aka phantom) accounts.
“Popular Anchorage restaurant targeted by scammers,” Alaska News Source, by Mike Mason, July 6, 2023.
“7 Common DoorDash Scams: How to Avoid Losing Out,” Ridester, by Brett Helling, March 31, 2023
© Copyright 2023 to Eric Niewoehner