Lyft teams up with nuTonomy to work on self-driving cars - Axios
Future of Work
Featured

Lyft teams up with nuTonomy to work on self-driving cars

While Uber is still entangled in court with Waymo, rival Lyft has inked a partnership with startup nuTonomy to further its own autonomous driving efforts, the companies announced on Tuesday.

Why it matters: Ride-hailing companies like Lyft and Uber have made bold statements about their futures are purveyors of self-driving-car rides. To make good on those statements, Lyft is inking new partnerships to develop the technology and software it'll need. Lyft already has a partnership with General Motors, although GM says it's not affected by the nuTonomy alliance.

The deal: The companies' work together will be focused on the passenger experience when hailing and riding in self-driving cars. After initially integrating nuTonomy's software and cars into Lyft's apps and service, the companies will deploy a pilot program in Boston, where nuTonomy is based, in the coming months. The test program will use Renault Zoe cars, just as with nuTonomy's other programs in Singapore and Boston, and a version of Lyft's app will be displayed on a console inside the cars.

The companies declined to comment on the number of cars the test program will initially include, though nuTonomy co-founder and CEO Karl Iagnemma said that his company wants "to end up with many thousands of nuTonomy vehicles on the Lyft platform in the future."

Who is nuTonomy: Founded in 2013 by MIT's Iagnemma and Emilio Frazzoli, nuTonomy is developing self-driving car software. The company began testing self-driving cars last year in Boston and Singapore (the latter in partnership with ride-hailing company Grab). To date, it has raised nearly $24 million in funding from Fontinalis Partners, Trucks Venture Capital, Highland Capital Partners, Signal Ventures, Bill Ford, and the government of Singapore, among others.

What about GM? In January 2016, General Motors announced a $500 million investment into Lyft and a partnership for the companies to work on self-driving cars togethers. "The announced partnership between nuTonomy and Lyft does not affect our existing business relationship with Lyft," a GM spokesperson told Axios via email. "Our plan remains steadfast — we are on track to launch our self-driving technology first in a ride-sharing application," but declined to provide timing details for rolling out a test program with Lyft.

Featured

Battery exec leaves Dyson two years after $90 million buyout

Michigan entrepreneur Ann Marie Sastry has left vacuum-maker Dyson, two years after it acquired her controverial lithium-ion battery company, Axios has learned. The $90 million all-cash buyout remains one of the richest lithium-ion deals ever.

Quick take: Sources with knowledge of the situation were not certain of the circumstances of Sastry's departure. But it comes eight months after Dyson relinquished Sakti3's core battery patents, and doubts remain in the field regarding her main claim, asserted repeatedly — that she was on the verge of commercializing much-sought-after solid state battery technology.

Why it matters: For the last two years, Dyson founder James Dyson has spoken of ambitious plans to spend $1 billion to $3 billion to revolutionize batteries and electric cars. He has said said his electric car will ready for the road by 2020. At the time, Dyson's October 2015 purchase of Sakti3 was the spearpoint of the mission, and Sastry's departure suggests more internal turmoil than he has let on.

  • Sastry's Linkedin page says she left Dyson last month. She identifies herself as the founder and CEO of a company called Amesite, which a source said is involved with artificial intelligence and education.

In September, Dyson told Bloomberg that he had created two competing battery teams—Sakti3, plus another that was attempting a different approach to solid state. One explanation for Sastry's departure was that the other team won. In an interview with the Guardian, Dyson said the company's batteries were already more efficient than those in commercial electric vehicles.

At the time of the October 2015 deal and since, numerous leading U.S. battery researchers told me they wondered why Dyson had bought Sakti3. Despite Sastry's robust claims of the company's progress with solid state, she had revealed very little publicly and, since no one else had made much progress, the deep suspicion was that she was exaggerating. Indeed, in reporting for a story at the time of the buyout, former Sakti3 executives told me that the doubters were correct—the company's technology was rudimentary and nowhere near commercial.

Dyson said Sastry is no longer with the company but declined to comment further. Sastry could not be reached.

Dan Primack contributed to this story.

Featured

Costco continues to hold its own against Amazon

Shares in Costco rose in after-hours trading, following the company's announcement of better-than-expected profits, combined with its 14th-straight month of same-store sales growth.

Costco's secret sauce is a mix of low prices, well-trained staff, and an ever-changing, but limited assortment of products, which have all kept Americans flocking to Costco outlets when other retailers have lost business to the Internet.

The most important number in Thursday's report was a 43.5% increase in e-commerce sales year-over year.

  • Though Costco stock is up more than 18% this year, it has lagged competitors like Amazon and Walmart over fears that the company is clinging too tightly to its profitable retail warehouses—these numbers will assuage some of those concerns.
  • During a call with analysts Thursday, CFO Richard Galanti stressed that Costco is experimenting with e-commerce "in our own way," and also, "pretty cheaply," using experiments like buy online and pick-up in store, which has the added benefit of driving traffic to its warehouses.
  • It's a delicate balancing act that, so far, investors are cheering.
Featured

AI advances at detecting cancer — but it can't see you now

Illustration: Rebecca Zisser / Axios

Medicine is poised to be one place where AI makes a mark. In a study published this week, researchers report that a machine algorithm was as good — or better — than pathologists at detecting the spread of a type of breast cancer.

For all the talk about the promise of AI radically changing medicine, this is one of the first peer-reviewed studies to back claims that algorithms can detect abnormalities in pathology slides, says Eric Topol from the Scripps Research Institute.

The bottom line: Radiologists and pathologists are likely to be the first in medicine affected by AI. But researchers working on the technologies don't see them replacing doctors, and instead aiding them. And even that role will require more data about the impact on the medical profession and whether AIs are accurate enough to diagnose patients.

“It is the early days,” Aidoc CEO Elad Walach says. “There’s not enough research at this point. Deep learning has been commoditized generally but it hasn’t been commoditized for the medical domain. The algorithms out there aren’t good enough as is. We need a lot of R&D to make AI work in this space. It is not just plug and play.”

What’s new: Babak Ehteshami Bejnordi and his colleagues from Radboud University Medical Center in the Netherlands evaluated algorithms submitted in a competition to analyze tissue samples from the lymph nodes of breast cancer patients. (Cancer cells are most likely to spread to these nearby areas first so they're involved in determining a patient's prognosis.) They then compared the accuracy of AI diagnoses with those of pathologists in two different situations where the researchers had a gold-standard test to check both:

  1. A panel of 11 pathologists had two hours to review 129 digitized images of samples from patients who had already received a diagnosis from a pathologist.
  2. One pathologist was given unlimited time to review all the cases. (The expert took 30 hours.)

The result: The top seven algorithms — all deep learning methods, which have lately seen progress in image and pattern recognition — performed better than the pathologists in identifying the metastases, but were on par with the pathologist whose time wasn’t restricted.

Keep in mind: The time constraints put — or not — on the pathologists in the study aren't the reality in which they practice. And, the AIs detected just one type of breast cancer. "We need to see it borne out across lots of other pathologies not just lymph nodes for breast cancers," says Topol. "This is the most impressive paper yet. But there are limitations. This is done in silico and is not a real world validation."

More opinions

PathAI: Andy Beck, whose team won the AI competition in the new study and who is now CEO of PathAI, says AI’s arrival to pathology will be a transformation rather than a disruption.

Seeing it as the latter “betrays a lack of understanding of how these fields operate. There are so many things physicians do. Typically an AI does one specific thing very well. We aren’t even close to doing the whole breadth of what a physician does.”

Aidoc: This Israeli startup is developing technology that can detect visual abnormalities — whether it be a cancer, stroke, bleeding or an edema — in head and neck CT scans. Their focus right now, says Walach, is on optimizing radiologists’ workflow from the current practice of reviewing cases in order received to getting AI to flag urgent ones first.

They’re currently testing their technology in 5 U.S. sites. Early, unpublished results at one hospital found that the AI could spot an abnormal scan with 98% sensitivity compared to what clinicians call the "ground truth" (in this study, the diagnosis by three radiologists working without AI), says Walach. They will look to publish their findings soon.

Walach says, “There is a need for peer reviewed publications about the outcomes not just the accuracy of these systems, and leading companies should invest time and resources in publishing clinical evidence.”

A big challenge: Like other cancer tests, there is a risk of detecting — and then treating — a cancer that isn't there. That isn't unique to machines but "algorithms are tuned to perform at maximum sensitivity, meaning there may be false positives," says Stanford University's Daniel Rubin, who develops imaging tools for radiology. “As we introduce these technologies, if people don’t improve accuracy and there are more false positives, it will increase the cost of health care.”

Go deeper: We asked four medical experts whether AI might help their profession

Featured

Walmart will start giving salary advances to 1.4 million workers

A worker helps a customer at a Walmart store. Photo: Julio Cortez / AP

Walmart announced Wednesday that it will start allowing its workers to claim already-earned wages before their scheduled payday, in order to help its associates meet unexpected expenses. It's working with FinTech firms Even and PayActiv to offer workers a suite of financial services, including the ability to take advances on pay up to 8 times per year.

Don't get too excited: Walmart has been investing in higher hourly pay for workers, but its starting wage of $9 per hour still trails that of rivals Target and Costco, and labor advocates say that allowing advances is little help for those who simply don't earn enough money.

Walmart tells the New York Times that the move will reduce financial stress for workers, which will make them happier and more productive, but critics are dubious.

  • "“It sounds like this may be a useful service but it doesn’t tackle the fundamental problem Walmart workers suffer,” said Paul Sonn, general counsel of the National Employment Law Project, tells the paper. “Their paychecks are too small.”
Featured

Zara's shrinking profit margins spooks investors

A Zara store in an upscale Istanbul neighbourhood in November 2017. Photo: Lefteris Pitarakis / AP

Inditex, the Spanish owner of Zara and largest apparel retailer in the world, announced a healthy 6% increase in net income for the first 9 months of 2017, but also a continued decline in profit margins — a measure that has been steadily falling since 2013.

  • Investors have bid the stock down the stock more than 2% during trading Thursday.

Why it matters: There is debate over whether the company can blame its profitability struggles on a strong Euro, but the 5% drop in Inditex stock this year shows that Wall Street's love affair with Inditex may be ending.

A best-in-class retailer: Despite the 2017 decline in Inditex's stock price, the firm has shown outstanding performance during the prior decade.

  • Since 2007, the firm has grown from roughly 3,100 stores to more than 7,500 today, on the strength of its decentralized supply chain and horizontally organized design team, which enables the firm to react to fashion trends faster than its peers.

Go deeper: Read the Wall Street Journal's piece.

Featured

AI pioneer Andrew Ng wants to modernize manufacturing

Ng says he likes being ahead of the curve. Photo: Eric Risberg / AP

After helping Google and Baidu kickstart their artificial intelligence efforts, Andrew Ng wants to see AI transform more than just tech companies. His new startup, Landing.ai, aims to help manufacturing companies tap AI to automate routine tasks, like inspecting products for defects.

The particulars: Ng says the new company is based in Palo Alto and has a couple dozen employees, but he wouldn't talk about funding or investors. It has a couple early customers, but the only one it is publicly talking about is electronics manufacturing giant Foxconn.

The back story: Ng ran Google's early AI effort, Google Brain, before helping China's Baidu ratchet up its AI effort. Ng told reporters that he likes being early to a space before it's conventional wisdom.

"I feel like in my life I've often done a lot of things before it was obvious it was a good thing to do," he said.

My thought bubble: Ng's chops in the field are second to none, but it's a bit unclear how a small team can effectively work with a large number of companies given how massive such operations tend to be. Ng says that while Landing.ai does custom work for each client, there are reusable elements. He says the company will be bigger a year from now than it is today.

Featured

Slow rebuild after the Great Recession killed economic equality

Apple's new campus in September. Photo: Marcio Jose Sanchez / AP

"Seattle is among a fistful of cities that have flourished in the 10 years since the Great Recession officially began in December 2007, even while most other large cities — and sizable swaths of rural America — have managed only modest recoveries," per AP Economics Writer Christopher Rugaber:

Why it matters: The rebound has "failed to narrow the country's deep regional economic disparities and in fact has worsened them."

More from the report:

  • "A few cities have grown much richer, thanks to their grip on an outsize share of lucrative tech jobs and soaring home prices. Others have thrived because of surging oil and gas production."
  • "In Las Vegas, half-finished housing developments, relics of the housing boom, pockmark the surrounding desert. Families there earn nearly 20 percent less, adjusted for inflation, than in 2007."
  • "[M]any Southern and Midwestern cities — from Greensboro, North Carolina, to Janesville, Wisconsin — have yet to recover from the loss of manufacturing jobs that have been automated out of existence or lost to competition from China."
Featured

Microsoft taps AI, Reddit to make Bing smarter

Microsoft AI and Research chief Harry Shum, speaking in San Francisco Wednesday. Photo: Ina Fried/Axios

Microsoft announced a bunch of new partnerships Wednesday as it aims to show itself as a leader in the field of artificial intelligence. It's also adding a bunch of AI-powered features to its own products, including its Bing search engine along with a deep integration of content from Reddit.

Why it matters: AI is one of the hottest areas in tech and Microsoft is competing with Google, Facebook, IBM and others for talent, mindshare and deals.

Partnerships: Microsoft announced efforts with a range of companies, including Reddit, UPS and China's Cheetah Mobile. In the Reddit deal, Microsoft's Bing search engine will use content from the online discussion community, including its popular "Ask Me Anything" Q&As.

Internal efforts

  • On the Bing front, Microsoft is using AI to deliver answers that combine information from multiple sites. That can allow results that compare different arguments on an issue, explore the differences between two things or just deliver a summary of facts from more than one place, with footnotes showing where the information came from. It is also making Bing more conversational and allowing people to search within images.
  • With Cortana, Microsoft is trying to help its digital assistant stand out from a crowded pack that includes Apple's Siri, Google's Assistant and Amazon's Alexa. Microsoft's case is that it is the only digital assistant that stretches across work and personal life. To that end, Microsoft said Cortana can now understand calendar and other data from Google's Gmail. It's also building Cortana into Android apps, including the CM Launcher app from China's Cheetah Mobile.
  • Microsoft is also expanding its use of AI in Office 365. A new insights feature will automatically make charts showing trends within a complex spreadsheet. Within word, AI will help make sense of acronyms within a document using other documents within the company. Microsoft already prioritizes which e-mails to read first, but a new feature will help find the action items within Outlook.
Featured

The debate over inequality

Last week, we reported that the wage inequality gap in the U.S., a primary source of the polarization among Americans, has been shrinking: For five straight quarters, wages have been growing the most for U.S. workers with only a high school diploma.

Data: Indeed analysis of BLS monthly jobs data; Chart: Axios Visuals

But readers pushed back:

  • "Surely you're kidding?" wrote James Harvey, executive director of the National Superintendents Roundtable. The percentage wage increase is better for high school graduates, he said, but the dollar increase still favors the rich: a 3.3% raise for someone making $20,000 a year is $660—only an eighth of the $5,000 raise going to someone earning $500,000 and getting a 1% increase.
  • In a phone call, Upwork CEO Stephane Kasriel told me, "The 1% is doing a lot better, and for the 30% and 40% at the bottom, it keeps getting worse."

Quick take: The times do indisputably favor the rich:

  • When adjusted for inflation, U.S. wages are up only 10% from almost a half-century ago.
  • Meanwhile, wealth held by the top 1% has surged: it rose to 38.6% of the total in 2016, from 36.3% in 2013, the Fed said in a report in September, while the bottom 90%'s wealth has fallen for almost three decades—last year, it was 22.8% of the total, compared with 33.2% in 1989.
  • FoW reader W.Spackman linked to this July report by Deloitte, which said income inequality today is comparable with the Gilded Age of robber barons, in the 19th century (Figure 3).

And Charlie Allenson, a FoW reader in New York, made another point: "Many of those chronically not working are not getting back to work. Namely, those more 'mature' workers. Ageism rolls on. Personal example: People look at my website and love my work. They meet me, see the gray hair and suddenly they're going in a 'different direction.' This is a constant for me. And it sucks."

But there are in fact signs of an improvement in the fortunes of ordinary people, and wages and salaries are among them, says Jed Kolko, chief economist at Indeed.com, the jobs listing website, who wrote the blog post on which we were reporting. In an email exchange, Kolko told me:

  • Harvey and Kasriel are correct to single out the vast concentration of income at the top, comprised largely of non-wage earnings like capital gains, interest and dividends—which combined are how the wealthy make most of their money.
  • But wages earned for work are another lens into the inequality story, and in that realm, the gap indeed has narrowed.
  • The shrinking wage gap is important to watch because wages and salaries are a large component of pre-tax money income, which includes interest, dividends and income from property. The Bureau of Labor Statistics puts the proportion at 76.8%. (see pages 8-9 in this BLS report).

The Fed, too, has noted the trend favoring less-educated workers. According to a Fed report released in September, income rose from 2013 through 2016 for all income groups, after accounting for inflation, which was a change from the prior three years, when income was stagnant. But the highest growth — an average of 25% — was among families without a high school diploma; in the 2010 to 2013 period, income fell for these workers, the Fed said.

Thought bubble: Inequality is not an absolute metric. If it were, ordinary people could legitimately lash out about wealth at the top regardless of how they themselves were faring. The rise of wages at the bottom and in the middle is slow, and the trend could halt—that is a point that Kolko makes. But it remains notable that the numbers are no longer going only in one, inexorable direction—there are metrics pointing to growing wages and salaries, and more jobs, for those whom the economy has been leaving behind.

Featured

Bitcoin: 'This is a casino, not an investment'

Illustration: Lazaro Gamio / Axios

Bitcoin is up about 1,700% since the start of the year. Some attribute the surge to ordinary, if enthusiastic, investment, along with the forces of supply and demand. Others say it's a bubble, and that it will ultimately burst. Joe Borg, president of the North American Securities Administrators Association, a grouping of state securities officials, suggests it's the latter. "This is a casino," he tells Axios, "not an investment."

The bottom line: Borg, who is also director of the Alabama Securities Commission, says he could be wrong and that those who say bitcoin is just "another type of investment" will be proven correct. But he sees worrying signs of a classic investment mania.

Among the signs:

  • People have told him they have taken out home equity lines of credit to buy bitcoin.
  • Those doing so, he said, are mostly millennials and young baby boomers.
  • "They seem to think anything electronic is a game," Borg said. "There are entrepreneurs who run Facebook, and they put this in the same category."

Thought bubble: If bitcoin collapses, which has been the normal course in big, sudden investment manias, the price is highly unlikely to go to zero, meaning a lot of people will still be in the money. But lots of people will lose, too, including perhaps some who have taken out those home equity lines of credit.

That there is a fever is indisputable. It is global, and especially heavy in Asia. Ordinary South Koreans are the most aggressive bitcoin investors, in addition to Hong Kong Chinese, Japanese and Vietnamese, report the WSJ's Steven Russolillo and Eun-Young Jeong. Together, they account for almost 80% of global bitcoin trading. Other reminders of fevers past:

  • Most of these Asians buying bitcoin are the general public, not professional traders.
  • At the point last week when bitcoin went above $17,000, it was almost $25,000 in South Korea, almost 50% higher, the WSJ said. In other words, the trade is chaotic to the point of irrationality.
  • At the FT, Izabella Kaminska writes today that even central banks are "getting drunk on the collective cryptocurrency/blockchain Kool-Aid."

A point that increasing numbers of observers are making is that bitcoin is only an investment, with no other real-life, large-scale utility, at least at present: bitcoin and other cryptocurrencies are too slow and cumbersome to serve as money, their original purpose.

  • In a speech today in Sydney, Phillip Lowe, the governor of the Reserve Bank of Australia, makes the point: "The current fascination with these currencies feels more like a speculative mania than it has to do with their use as an efficient and convenient form of electronic payment."