As the world moves toward a computer-dependent living, it comes as no surprise that many new companies have sprouted, advertising themselves to have a scalable AI technology, even if there's not. But why is this happening?
For a start, building an effective artificial intelligence technology isn't a cakewalk. In fact, it requires a raft of data, time, and money, which means a seriously massive upfront investment before materializing a bare minimum program.
This is why some AI startups resort to hiring humans as a front for AI bots, so companies that have meager resources will be able to peddle themselves to potential investors to have an existing AI tool. For Gregory Koberger, the chief executive officer of ReadMe, there are many such startups.
"Using a human to do the job lets you skip over a load of technical and business development challenges. It doesn't scale, obviously, but it allows you to build something and skip the hard part early on," Koberger told The Guardian, adding that "it's essentially prototyping the AI with human beings."
How to start an AI startup— Gregory Koberger (@gkoberger) March 1, 2016
1. Hire a bunch of minimum wage humans to pretend to be AI pretending to be human
2. Wait for AI to be invented
But other than securing financial backing from investors, hundreds, if not thousands of companies, are faking it due to privacy concerns. It has been proven that customers are more trusting and open to AI bots than human counterparts.
In 2008, voicemail-to-SMS firm Spinvox was reportedly using humans in offshore call centers to transcribe voicemails. In 2016, calendar scheduling platforms Clara and X.ai were busted using humans to act as chatbots.
In 2017, expense management app Expensify claimed to have a "smartscan technology" to process receipts of customers but later admitted that they were using humans to transcribe data. Just recently, it was discovered that email app Edison Software was using humans to scan and read email messages to develop a "smart reply" feature.
There's no surprise here that using humans for the work of a computer isn't just money-motivated because even Facebook, a multibillion-dollar company and a huge spender on AI research, is relying heavily on humans for its digital assistant for Messenger, M.
While consumers and policymakers worldwide are demanding for user privacy to be respected, at this point, it remains uncertain how long will companies be transparent with their service.