Juno was supposed to be the antidote to Uber and Lyft and all the other gig-economy driving platforms. The startup promised to limit its commissions and share equity with its drivers. Instead of doing piecework for a software-driven boss, you’d actually own a tiny piece of the company that parceled out rides to you.
But now Juno has been acquired by another ride-hailing outfit, the Tel Aviv-based Gett, for $200 million (Johana Bhuiyan in Recode). As a result, it is liquidating its equity-sharing model. Drivers who’d worked enough to qualify are being paid off on the order of two cents a share, receiving payouts in the vicinity of $100.
The new machine-learning techniques transforming how digital systems operate don’t work like old software programs. Once upon a time, programmers wrote code, and the code worked — or it didn’t, and the programmers debugged. But how do you debug a neural network? In Technology Review, Will Knight explores “the dark secret at the heart of AI”: We can’t really know why AIs do what they do. They are not programmed, but trained.
When a machine-learning program provides an answer, a decision, or a choice that its human operators believe is wrong, they can tell it so, and it will incorporate that data into its next choices. But most AIs can’t turn around and tell us how they reached a particular outcome. This “explainability” problem poses practical, legal, and moral questions we’re only beginning to scope out.
When it comes to experimenting with contractors’ behavior, how far do gig-economy companies like Uber go? Pretty far, according to a new in-depth account by The New York Times’ Noam Scheiber.
In a sense, the entire Uber platform is like a Skinner box or ant farm for isolating and examining how different incentives and inputs produce different outputs from riders and drivers. Uber is constantly tweaking the behavior of its drivers, the people formerly known as employees, with gamification-style interface elements — like badge rewards or notifications that pop up to say, “You’re $10 away from making $330 in net earnings. Are you sure you want to go offline?”
A central player in a deepening scandal has decided to “take the Fifth” — to remove himself from having to testify in a court proceeding because it could incriminate himself.
No, we’re not talking about the Trump administration’s Russia troubles. This is the latest twist in the legal battle between Google and Otto’s Anthony Levandowski, the self-driving truck company founder whose firm was acquired by Uber last year. Google has accused Levandowski of stealing its autonomous vehicle tech when he decamped from its subsidiary Waymo to Uber, and now Levandowski has taken the advice lawyers so often give and chosen to remain silent (TechCrunch).
Offices are old hat. Distributed teams and remote work are the wave of the future. Right? Certainly, the numbers have been moving in that direction. Today, big companies seek to hire the best talent wherever it might live, while small firms skip the “rent an office” step and spend the cash on team gatherings and retreats. Employees opt for flexible schedules over painful commutes. And everyone ends up more productive.
But every trend has a counter-trend. IBM, once a distributed-workforce pioneer, is moving in the other direction (Sarah Kessler in Quartz). In the latest phase of a wave of “co-location” efforts, thousands of IBM marketing employees have been told to stop working from home and report for duty at one of six central offices.
When the first Trump travel ban rolled out in a flurry of airport chaos and outraged protests last month, tech companies flocked to join the swell of opposition. 127 firms signed on to a friend-of-the-court brief supporting the legal challenge to the president’s executive order that led to its demise.
But the arrival of Trump Ban 2.0 this week has evoked less of a response from tech, and so far, although 58 companies have signed on to support a challenge to the new order, Google, Apple, and Facebook are not among them (Reuters).
It’s hard to believe things could get even worse for Uber after the past week’s gender discrimination allegations, lawsuits, and resignations. But get worse they did. Yesterday Bloomberg published a video, secretly captured by a dashcam, in which CEO Travis Kalanick gets into an argument with his Uber driver, Fawzi Kamel. Kamel complains that Uber’s price drops have bankrupted him; Kalanick responds, “Some people don’t like to take responsibility for their own shit.”
Being CEO of a billion-dollar company is kind of like being president: You’d better assume you’re always on camera. Kalanick issued what he called “a profound apology” in an email to his company, writing, “I must fundamentally change as a leader and grow up. This is the first time I’ve been willing to admit that I need leadership help and I intend to get it.” (Paging Jerry Colonna.)
Yes, it is absolutely true that many forms of harassment against women in the workplace have been going on forever, and yes it is true that sometimes, the individuals responsible for said harassment go unpunished. And it is also true that there are many bro-grammer cultures that can’t seem to figure out how to treat all employees fair and equally.
Engineer Susan Fowler set off a firestorm last week with a post describing her experiences during a year’s employment at Uber. It was hard to tell which was worse to read about — the incidents of sexual harassment she described, or the inadequate (or nonexistent) response to her treatment by the firm’s management and human-resources team.
Uber, of course, has always been known for a brash, bare-knuckles culture. But Fowler’s account somehow broke through the industry’s indifference with telling, all-too-credible detail. Maybe it was the immunity seemingly granted “high performing” managers to hit on, condescend to, and discriminate against female colleagues; maybe it was the petty stuff, like the time everyone on the engineering team got status-conferring leather jackets, except the women, because theirs would cost a little more.