Now eight months on the job, Uber CEO Dara Khosrowshahi has one of the tallest – and most closely scrutinized – orders in tech: fix the much-maligned company culture, bring together a fractured board, grow, and go public in the next year. The New Yorker’s profile details how Khosrowshahi is updating the executive team and addressing driver unrest, and what’s next for Uber’s autonomous vehicle program in light of a fatal pedestrian accident in Arizona:
When I reached Khosrowshahi by phone, he seemed disheartened, and disarmed by the intense scrutiny that comes with his new job. He told me that the autonomous division had been working toward offering driverless-car service by the end of the year, and that there would inevitably be “bumps and bruises” along the way. “What happened last week was truly tragic,” he said. “We’ve clearly taken a very, very big step back.”
As University of Toronto professor and entrepreneur Ajay Agrawal has noted, AI can be framed as a drop in prediction costs, accompanied by a relative increase in the value of human judgment. Per MIT’s Andrew McAfee, AI is rapidly encroaching on the latter, with profound implications for the C-suite:
A lot of executives think that a big part of their job is making the tough calls; relying on the experience, industry knowledge, and judgment that they’ve built up … [But] over and over again we’re seeing that technology is better at human-judgment tasks than humans. To me, it’s the most unsettling by-product of the machine-learning revolution.
The Economist, in its series on AI in the workplace, outlines what this disruption looks like in HR, where AI tools are changing the way companies recruit, hire, and manage talent:
Twine Labs, a startup that is working with Nielsen, suggests internal candidates for new roles, based on employee data and job requirements, taking in hundreds of variables. Around half the candidates it suggests are approved and promoted, says Joseph Quan, Twine Labs’ boss. That is about the same success rate as for a human recruiter.
Increasing your company’s R&D investment may be in order. Though Amazon doesn’t report R&D specifically, Bloomberg’s back of the envelope calculation puts the company’s annual expenditure at $22 billion, a whopping $6 billion over Alphabet, the next biggest spender:
Can France become the most exciting country for technology innovation? President Emmanuel Macron recently announced his country’s new AI strategy, including a €1.5 billion investment in establishing France as a machine learning hub alongside the US and China. In Wired, Macron explains how his government will ensure accountability and transparency in AI:
We will open data from government, publicly funded projects … and we will favor, incentivize the private players to make [their algorithms] totally public and transparent. Obviously some of them will say, there is a commercial value in my algorithm, I don’t want to make it transparent. But I think we need a fair discussion between service providers and consumers, who are also citizens and will say: “I have to better understand your algorithm and be sure that this is trustworthy.”
Foreign Policy goes to Rongcheng, where a pilot for China’s national social credit program scores residents on everything from obeying traffic laws to charity donations, even granting points for committing a heroic act. Businesses are not exempt from this system of enforced “trustworthiness”:
Companies are also included in the gauntlet of social credit. They can remain in good standing if they pay taxes on time and avoid fines for things such as substandard or unsanitary products — a sore point for Chinese people, who tend to mistrust firms and service providers due to frequent scams and food safety scandals. High-scoring businesses pass through fewer hoops in public tenders and get better loan conditions.
Last year’s Equifax breach plunged credit rating agencies into the spotlight for cybersecurity concerns, but this overlooks deeper questions about how computers changed estimates of credit worthiness over time. A historical perspective explains why this matters to the viability of these widely used but often misunderstood institutions:
The credit bureaus wanted to collect information about people’s personalities, and about their home lives and all of that sort of stuff, which was not really compatible with computers because back then computers didn’t have the same kind of memory or processing they have now. So they had to reduce the categories they used to determine creditworthiness. I think that was antithetical to the idea of collecting as much information as possible from any source possible.
Women who have their first child during prime career-building years (between age 25 and 35) never recover the pay gap with their partners, while those who become first-time mothers outside of that range do. Surprisingly, this pattern holds in Scandinavia, not just the US, according to a study in The New York Times:
“The birth of a child is really when the gender earnings gap really grows,” said Danielle H. Sandler, a senior economist at the Census Bureau and an author of the paper. The study found that over all, women earn $12,600 less than men before children are born and $25,100 less afterward. … The pay gap grows larger with each additional child. It does not begin to shrink until children are around 10.
