Dating Sites Have Fake Babes: 5 Ways Tech Companies Rip Us Off
The tech industry prides itself on being at the forefront of innovation. Unfortunately, a lot of what they're innovating is new ways to rip off users. In fact, there are whole areas of tech that seem to be infested with more scammers than your parents' hotmail inbox.
There Was A Second Theranos-Style Scandal (Involving Poop)
Every Silicon Valley success sparks imitators and Theranos was no exception, although in their case it was more of a copycat crime. Like Theranos, Ubiome was a hot health tech startup that raised millions in VC funding. While Theranos was going to revolutionize blood testing, Ubiome promised to do the same for fecal testing with an affordable at-home test that would check your gut health. But after Theranos collapsed, the "Theranos of poop-testing" turned out to be the ... Theranos of poop-testing. In that they were both weird scams.
While Theranos was just faking their tech and dumping all your blood down a Starbucks toilet, Ubiome was secretly billing customers' insurance providers up to five times for the same shitty test. And we mean "shitty" in multiple ways there, since the tests often had no medical value for patients. You'd be better off taking a jar of your shit to Gwyneth Paltrow's unlicensed fecal psychic. She might start rambling about Buddha's secret stool softener, but at least she'd only charge once for it.
Because insurance companies require tests to be ordered by a doctor, Ubiome hired an in-house team of doctors who were instructed to approve all tests, regardless of whether they were medically useful. At least one doctor was fired for not rubber-stamping tests fast enough. Meanwhile, the company was using stock photos in their "testimonials," as well as a bunch of other shady advertising tactics. When insurance companies began rejecting their claims, they started only informing investors of "billable" claims rather than approved ones, making growth look rapid. It's like a restaurant claiming to have thousands of customers, even though everyone immediately spat out the food and ran out without paying.
The Theranos comparisons don't stop there. Founders Jessica Richman and Zac Apte were secretly in a relationship, like Theranos founders Elizabeth Holmes and Sunny Balwani. Richman also had a bunch of weird, Holmes-like personal quirks, including lying about her age to get on "founders under 30" lists (she was 45) and claiming without evidence to have been a former bodybuilder, which is a crime usually reserved for very horny men trying to bring a bottle of champagne into LA Fitness. (Quite unsurprisingly, the FBI raided the place in 2019, launching a major investigation.)
Data Security Firms Mostly Just Pay Off Your Hackers
Businesses and governments face a growing problem with ransomware attacks, in which hackers encrypt data or block access to crucial systems and then demand a ransom to fix the problem. Luckily, data recovery firms like Proven Data and MonsterCloud are here to help. These "ransomware experts" promise to decrypt files quickly and effectively. In return, they charge a hefty fee, usually larger than the actual ransom. But customers were understandably willing to pay a little more to resolve the situation the "right" way, without making some shady deal with the criminals. Except that Proven Data and MonsterCloud would just pay the ransom and keep the rest of the money.
This is a huge problem in cybersecurity. One researcher secretly created his own ransomware and sent requests for help to a number of security firms. They all claimed to be experts in decrypting that particular ransomware and then quietly inquired about paying the ransom. It's apparently pretty easy money. Only one of Proven Data's founders had an IT background, and he rarely showed up for work because he was more focused on his side gig as a DJ. It's very dispiriting to hire an elite tech expert to battle hackers, only to later see him being thrown out of Steve Aoki's party bus for doing too many whip-its and trying to eat an MTV Award.
There are legitimate reasons for a security company to just shrug and pay the ransom. Movies and TV have convinced people that decryption is just a matter of finding a hot nerd to type really fast at the problem, but it's actually very hard! And hackers often target hospitals and medical firms, where any delays in decryption can mean life or death. As a result, there are cybersecurity firms who are open about paying ransoms sometimes. But Proven Data and MonsterCloud originally presented themselves as expert decrypters who would use every tactic to bring systems back online. "Every tactic" of course includes paying ransoms, which was actually the "standard procedure."
And you really don't want paying ransoms to be standard procedure. For starters, it can encourage hackers to strike again, because they know they'll just get another payday. And the firms can have a mutually beneficial relationship with hackers. When the SamSam ransomware attacked systems across the US, Proven Data paid so many ransoms that they kept a regular channel open to the hackers, who would agree to keep compromised systems running while Proven Data negotiated their contract. In return, the SamSam hackers actually started recommending that their victims use Proven Data. If a guy mugs you and then on the way out shouts "Oh no, I hope I didn't drop a business card for famed private detective Slade Renoir!" maybe find someone else to get your phone back.
Cloud-Based Medical Records Are Apparently A Scammer's Paradise
Until recently, medical records were simply inscribed on clay tablets and then buried in a swamp to appease the angry gods. But we have the Internet now and the gods can't get any angrier, so it makes sense to move those records online. It's become a huge field in Silicon Valley, with startups raising millions by promising to make medical records accessible on the Cloud. This would be a huge help to doctors, if only the companies didn't keep turning out to be evil. And not, "overcharged you" evil. We're talking "everyone who works there has blood-red eyes and speaks only Aramaic" evil.
Take Practice Fusion, which promised to digitize medical records on the cloud, while using cutting edge data analysis to help doctors detect problems and design treatment plans. The company raised over $150 million in funding and was planning a $1.5 billion IPO in 2016. Then in 2018, it suddenly sold itself for $100 million in a very shady deal that left employees and stockholders with nothing but made executives millions. What happened? Well it turned out that Practice Fusion had been taking kickbacks (sorry, "sponsorships") from pharmaceutical companies. In exchange, its software would recommend their drugs to doctors even if they weren't medically necessary for the patient.
For example, an opioid manufacturer paid the startup to create a pop-up asking about a patient's pain level, then a dropdown menu would recommend various treatments, including "opioid therapy" using the company's drugs. The program did this even in cases where opioids were against medical guidelines. Practice Fusion promised that opioid makers could expect an extra 2,777 patients per doctor with their software, most of whom had no need of highly addictive opioids. The company reached a $145 million settlement with the Justice Department last month, because pushing dangerous drugs only really deserves prison time if you're doing it out of the trunk like a gross old Fiat or something.
Practice Fusion's big rival for hottest health records startup was a company called eClinicalWorks (ECW), whose software is used by major clients like the NY Department of Corrections. Unfortunately said software was a buggy piece of shit that regularly lost patient history, recommended inappropriate or defunct drugs, didn't transmit prescriptions, displayed one patient's name and another's medical history ... just the worst stuff imaginable. ECW deliberately concealed all these problems from the regulatory body to get government certification. They also bribed doctors with tens of thousands of dollars to provide public endorsements, which were then used to obtain certification.
The software was so bad that it seems to have literally killed people. In one case, a doctor ordered a brain scan, but ECW's software never sent that request over to the lab. The patient died of an undetected aneurysm a short time later. Meanwhile prisoners on Rikers Island were getting totally messed up because the system was recommending the wrong medication. The company agreed to a $155 million settlement with the Justice Department in 2018. And it's nice that the Justice Department is so caked up now, but the government could probably take a stronger stance than "hey, why not kick a little of that cash our way?"
Groundbreaking AI Tech Keeps Turning Out To Be Low-Paid Human Contractors
Artificial intelligence is one of the hottest things in tech right now, with AI startups raking in an astonishing $31 billion by the end of 2018. But true AI is difficult to exploit and expensive to build, whereas old-fashioned humans are easy to exploit and built for free, as long as everyone's drunk enough. Which is why a number of "AI" startups have turned out to just be a front for a bunch of low-paid employees secretly doing all the work. That's right, in the real world Skynet was just an office building full of temp workers engaged in some very aggressive contract negotiations.
For example, a startup called Engineer.ai became a hot investment after its CEO and "Chief Wizard" claimed that the company's AI could help users design and code mobile apps. Engineer.ai assured users and investors that its AI was "80% done" and that it just used a few human engineers to fill in the gaps for the time being. But the Wall Street Journal reported that the company was actually just hiring Indian engineers to do all the design work. At best, they had vague plans to maybe start using AI at some point in the future. In fact, the WSJ could find no evidence of "any kind of AI agent or software of any kind" building the apps. There's more artificial intelligence in a Magic 8-Ball.
Scheduling services like Xan.ai and Clara actually relied on workers spending 12 hours a day pretending to be chatbots. Expensify promised to use proprietary "smartscan" tech to process your receipts when it was just sticking them on MTurk for gig workers to transcribe. GoButler claimed to be an "AI-assisted" concierge service, where you could text a request for anything from pizza to an antique human skull and the cutting edge tech would arrange everything. In reality, the company launched with no AI at all, all those texts were just going to a team of overworked call center employees (their official job titles were stuff like "hero" and "superhero") who were so overwhelmed by the constant demands they had to attend their own Christmas party in 30 minute shifts.
Even big companies aren't entirely immune. Facebook's discontinued AI virtual assistant "M" relied heavily on humans to answer requests. There's also a disturbing new trend for "robots" that are actually piloted by low-paid workers, often in the developing world. Berkeley used to have adorable, burrito-delivering robots rolling around campus, which were actually controlled by Colombians earning less than $2 per hour. A Japanese company even wants to introduce robot maids that are actually controlled by remote workers. The plan is to eventually outsource the controller job, allowing busy families to get a little help around the house without having to pay a living wage or let any foreign people into the country. Plus, all those workers will get to enjoy the wonders of the first world through the robot's eyes. Nothing dystopian there!
Dating Sites Keep Using Fake Accounts To Keep You Signed Up
Using fake accounts to lure customers is a time-honored tradition in Silicon Valley. When Reddit launched, the founders added a special tool allowing them to quickly submit links and comments under a variety of fake profiles, making the site look like it had way more users than it actually did. They now openly brag about this in speeches, suggesting that the practice isn't exactly frowned upon. Admittedly, it's a much bigger problem when we're talking about paid accounts, rather than a free-to-use site like Reddit, but it turns out that's the exact model embraced by many of the biggest players in online dating.
Many dating sites allow free accounts to receive messages, but ask users to upgrade to a paid account to reply. The basic version of the scam is for a site to create fake accounts to message free users, enticing them to sign up for a paid plan. For example, JDI Dating, which boasts millions of paid accounts on sites like findmelove.com, was fined by the FTC for using computer-generated fake profiles to target new users. But JDI were mere carnival goldfish in the giant sewer of matchmaking evil. And the alligators were finding even worse ways to target their most vulnerable users.
Take Global Personals, the UK giant that runs thousands of branded dating sites worldwide (if your local paper has a personals site, it's probably run by Global Personals). The company hired a team of employees called "pseudos," who would analyze the profiles that got the most responses, then steal their photos to set up fake accounts in other markets. Each pseudo ran at least 15 fake accounts, which were used to send messages enticing unsuccessful users to stay signed up for a monthly paid account. Some users spent up to 24 months messaging with the same fake account, while others were talking to multiple fakes run by the same Global employee. If a user became too insistent about meeting, the pseudo would break off contact, then just send new messages under a different identity.
But the most sophisticated version was being run by market leader Match.com. Unlike most of their users, Match understands the importance of keeping their hands clean. So they didn't set up any fake accounts directly. Instead, they only notified users about spammers and fraudulent accounts if they had a paid account. Literally hundreds of thousands of users received what they thought was a real message and signed up for the paid account to write back. It was only then that Match would inform them the account was actually a scammer. According to the FTC, Match refused all refunds for this and deliberately made it difficult to cancel the paid account. Courtship in the 21st century, folks!
Top image: Tero Vesalainen/Shutterstock
