Machine learning is a cost reduction multiplier for staffing costs, which are perhaps the highest cost for technology firms until they become successful. Googl

Machine learning is a cost reduction multiplier for staffing costs, which are perhaps the highest cost for technology firms until they become successful. Google and Facebook use algorithms to try and minimize dollars spent on moderating their sites, with limited success. If they can avoid paying human beings to make judgement calls, they save billions of dollars a year. It’s reasonable to be skeptical of startups who claim that they’ll use ML someday. Ask them how they’re using human labor today to perform the tasks that they’d like ML to do, and what their runway is for that work at their current burn rate. If they don’t deliver a viable ML labor reduction by that time, they will either collapse or worsen their service.