Year: 2018

Over the past few years, Chinese automaker giant BYD has been on a partnership spree with cities across China to electrify their public transportation systems, and now it’s extending its footprint across the globe. On Thursday, BYD announced that it has shipped 100 electric buses to Santiago, the capital city of Chile.

The step marks the Chinese firm’s further inroads into the Latin American country where a green car revolution is underway to battle smog. BYD’s first batch of vehicles arrived in Santiago last November and the Warren Buffett-backed carmaker remains as the only electric public bus provider in the country.

Chile is on the map of China’s grand Belt and Road Initiative aiming to turbocharge the world’s less developed regions with infrastructure development and investments. “With the help of ‘One Belt One Road,’ BYD has successfully entered Chile, Colombia, Ecuador, Brazil, Uraguay and other Latin American countries. As the region accelerates its electric revolution, BYD may be able to win more opportunities,” said BYD in a statement.

President of Chile Mr. Sebastián Piñera rides the BYD electric bus. / Credit: BYD

The 100 buses embarked on a 45-day sea voyage from BYD’s factory in eastern China to land on the roads of Santiago. They sport the Chilean national colors of red and white on the exterior and provide USB charging ports inside to serve a generation who live on their electronic devices.

The fleet arrived through a partnership between BYD and Enel, a European utility juggernaut that claims to make up 40 percent of Chile’s energy sales in 2017. Enel has purchased the fleet from BYD and leased them to local transportation operator Metbus while the Chilean government set the rules and standards for the buses, a BYD spokesperson told TechCrunch.

Local passengers graded BYD’s electric vehicles at 6.3 out of 7, well above the 4.6 average of the Santiago public transportation system, according to a survey jointly produced by Chile’s Ministry of Energy as well as Ministry of Transport and Telecommunications. Respondents cited qualities such as low noise level, air conditioning and USB charging that the buses deliver.

Santiago currently has 7,000 public buses running on the road, among which 400 get replaced every year. A lot of the new ones will be diesel-free as the Chilean government said it aims to increase the total number of electric vehicles by tenfold in 2022.

Responding to criticism from his recent trip to Myanmar, Twitter CEO Jack Dorsey said he’s keen to learn about the country’s racial tension and human rights atrocities, but it has emerged that both he and Twitter’s public policy team ignored an opportunity to connect with a key civic group in the country.

A loose group of six companies in Myanmar has engaged with Facebook in a bid to help improve the situation around usage of its services in the country — often with frustrating results — and key members of that alliance, including Omidyar-backed accelerator firm Phandeeyar, contacted Dorsey via Twitter DM and emailed the company’s public policy contacts when they learned that the CEO was visiting Myanmar.

The plan was to arrange a forum to discuss the social media concerns in Myanmar to help Dorsey gain an understanding of life on the ground in one of the world’s fastest-growing internet markets.

“The Myanmar tech community was all excited, and wondering where he was going,” Jes Kaliebe Petersen, the Phandeeyar CEO, told TechCrunch in an interview. “We wondered: ‘Can we get him in a room, maybe at a public event, and talk about technology in Myanmar or social media, whatever he is happy with?'”

The DMs went unread. In a response to the email, a Twitter staff member told the group that Dorsey was visiting the country strictly on personal time with no plans for business. The Myanmar-based group responded with an offer to set up a remote, phone-based briefing for Twitter’s public policy team with the ultimate goal of getting information to Dorsey and key executives, but that email went unanswered.

When we contacted Twitter, a spokesperson initially pointed us to a tweet from Dorsey in which he said: “I had no conversations with the government or NGOs during my trip.”

However, within two hours of our inquiry, a member of Twitter’s team responded to the group’s email in an effort to restart the conversation and set up a phone meeting in January.

“We’ve been in discussions with the group prior to your outreach,” a Twitter spokesperson told TechCrunch in a subsequent email exchange.

That statement is incorrect.

Still, on the bright side, it appears that the group may get an opportunity to brief Twitter on its concerns on social media usage in the country after all.

The micro-blogging service isn’t as well-used in Myanmar as Facebook, which has some 20 million monthly users and is practically the de facto internet, but there have been concerns in Myanmar. For one thing, there was been the development of a somewhat sinister bot army in Myanmar and other parts of Southeast Asia, while it remains a key platform for influencers and thought-leaders.

“[Dorsey is] the head of a social media company and, given the massive issues here in Myanmar, I think it’s irresponsible of him to not address that,” Petersen told TechCrunch.

“Twitter isn’t as widely used as Facebook but that doesn’t mean it doesn’t have concerns happening with it,” he added. “As we’d tell Facebook or any large tech company with a prominent presence in Myanmar, it’s important to spend time on the ground like they’d do in any other market where they have a substantial presence.”

The UN has concluded that Facebook plays a “determining” role in accelerating ethnic violence in Myanmar. While Facebook has tried to address the issues, it hasn’t committed to opening an office in the country and it released a key report on the situation on the eve of the U.S. mid-term elections, a strategy that appeared designed to deflect attention from the findings. All of which suggests that it isn’t really serious about Myanmar.

TransferGo, the London-based international money transfer startup, has raised just over $17.6 million in Series B funding, including an earlier tranche of funding closed in May. The round is led by Vostok Emerging Finance, and Silicon Valley’s Hard Yaka, with participation from Revo Capital, U-Start Club, and Practica Capital. The figure also includes around $830,000 in equity crowdfunding via Seedrs.

Founded in 2012, TransferGo currently operates in 47 countries around the world, with offices in London, Vilnius, Berlin, Warsaw and Istanbul. It claims a customer base of 833,000 users, adding more than 1,000 new customers per day, and positions itself as offering one of the fastest international money transfer services on the market. This sees it able to provide international “cross-network” transfers in 30 minutes.

The fintech company also recently launched a free international money transfer service, with what it says is a a zero transaction fee and with no mark up on exchange rates, allowing customers to transfer money globally at no cost. Essentially, if you aren’t time sensitive or perhaps are transferring larger amounts, you can elect to use the free tier. If you need a guaranteed arrival time for the money you are sending, you can use the paid tier, which still looks pretty competitive.

In a call and over follow-up emails, TransferGo co-founder and CEO Daumantas Dvilinskas explained that the fintech has built out its own “proprietary technology and infrastructure” to enable it to do 30 minute transfers on the corridors it offers and at a cost that remains low. This means having partner bank accounts as close to the final destination as possible, and re-routing the money being transferred to avoid unnecessary charges and to enough volume to afford economies of scale at the point of conversation.

“We cross-sell our customers different delivery options based on how quickly they want to receive the money,” Dvilinskas tells me. “TransferGo Now product, where customers can get a 30 minute guaranteed, together with other speedy delivery options, effectively pays for the TransferGo Free product… At the same time, economies of scale have been decreasing our direct cost of transactions to a point when we can offer the free product in a sustainable fashion”.

Dvilinskas says TransferGo’s typical customer is “a global citizen” who receives a salary abroad and sends around $500 back home every month. “Before using us they would have been using a cash bureau, bank or PayPal. In addition to this core segment, we see a growing larger transaction segment who are leveraging our competitive currency conversion rate for larger transactions to pay bills or buy goods abroad,” he adds.

Historically, TransferGo launched to enable people in the U.K. to send money to Central and Eastern Europe, a corridor where it claims 20 percent market share (based on the World Bank data). However, the fastest growing corridors today are Continental Europe to Ukraine, Turkey, India, and other emerging market destinations.

“Specifically, Poland, Germany, and Turkey are emerging as important send markets, which is where we opened up offices last year,” says Dvilinskas.

Netflix loves to test new ideas, and its latest experiment is an odd new feature that lets viewers watch a scene again.

A selection of Netflix subscribers noticed the new addition, which serves a pop-up asking if they want to “watch this scene again” after certain ‘highlight’ scenes in a show.

The streaming giant confirmed the pilot to the Los Angeles Times, adding:

We’re trying out a feature which gives Netflix members the ability to rewatch favorite scenes and memorable moments with the click of a button. Right now we’re just looking to learn from it and may or may not roll it out more broadly in the future.

I can’t say I’ve ever had the urge to watch a scene again — and I spend a considerable amount of time on Netflix, often with kids — so this is a pretty curious test.

As you might imagine, early users haven’t been too impressed either. One anonymous subscriber took to Reddit to bemoan how it “devalued” the film they were watching. That person was watching ‘Dumplin,’ but even still it isn’t hard to imagine how frustrating multiple popups would be.

Other Netflix tests from the past have included video promos between episodes, and showing shows on the log-in screen. On the business side, it has experimented with bypassing in-app subscriptions and also a new mobile-only package to make its service more affordable in emerging markets.

But experimentation and thinking differently is often a key part of what makes a business successful and Netflix certainly knows a lot about the latter.

The company just broke new records on consumer spending in its mobile apps during November, according to data from app intelligence firm Sensor Tower. It is said to have grossed $86.6 million during the month, a whopping 77 percent annual rise, with increasing revenue coming to Netflix from its international markets.

The funding extravaganza may be approaching its end for scooter “unicorns” Lime and Bird, but smaller startups in the micro-mobility space have continued to close venture capital rounds at a consistent pace. See Grin, Tier and Yellow for examples.

The latest is Dott, a European scooter startup founded by Maxim Romain, Ofo’s former head of Europe, the Middle East and Africa. Romain joined Ofo, a Chinese bike- and scooter-sharing company that raised more than $1 billion in venture capital funding but has struggled to scale overseas, in 2018 to help it expand. He only lasted seven months before realizing he could do it better himself.

“Why work for a Chinese company when we can do it ourselves in Europe where we better understand the market?” Romain told TechCrunch. 

Dott, headquartered in Amsterdam, has raised €20 million in a round co-led by EQT Ventures and Naspers. Axel Springer Digital Ventures, DN Capital, Felix Capital and others also joined. Dott is using the capital to launch in several cities across Europe, beginning with an early 2019 e-scooter pilot at Station F, a startup campus located in Paris. Additional launches are in the pipeline, as are electric bikes.

As a result of its learnings from Ofo, Bird and Lime, all of which have struggled to keep their equipment out of disadvantageous spots, like trees, lakes and garbage cans, Dott says it’s built sturdier scooters. They have 10” wheels, wider decks, a double brake system for safety, a speed cap at 20km/h and apparently are able to hold a charge longer than competing scooters — though we couldn’t independently verify this.

Dott says it’s taking a friendlier approach to launching in new cities, again, unlike some of its predecessors. If you remember, Bird showed up in a number of cities without permission — a move that resulted in it being denied a permit to operate in San Francisco. Dott will hire local teams to collaborate with city officials to develop pilot plans tailored to each market and it won’t rely on gig economy workers to recharge, clean and maintain scooters. Instead, it will hire and train a team of Dott employees dedicated to maintenance in each city.

“I think a lot of the companies grow too fast in the sense that they don’t necessarily have the product that can enable them to be profitable but because they want to win the race,” Romain said. “They want to raise as much money as possible as quick as possible and to deploy scooters as quick as possible. This creates an environment for them where their unit economics are extremely bad.”

“That’s exactly what we saw with bike-sharing in China. In the end, the reality of the unit economics came back to bite them. It’s a risk. Lime and Bird are doing a lot to improve their hardware but it’s a risk for the industry. For us, we are taking the view that we really need to focus on the product so we have the right unit economics and we can be sustainable. If you want to make it happen, you have to make it happen in a sustainable way.”

LemonBox, a Chinese e-commerce startup that imports vitamins and health products from the US, has raised $2 million to develop its business.

The company graduated from Y Combinator’s most recent program in the U.S. and, fuelled by the demo day, it has pulled in the new capital from 10 investors which include Partech, Tekton Ventures, Cathexis Ventures, Scrum Ventures and 122 West Ventures.

LemonBox started when co-founder and CEO Derek Weng, a former employee at Walmart in the U.S, saw an opportunity to organize the common practice of bringing health products back in China. Any Mainland Chinese person who has lived or even just visited the U.S. will be familiar with such requests from family and friends, and LemonBox aims to make it possible for anyone in China to get U.S-quality products without relying on a mule.

The service is primarily a WeChat app — which taps into China’s ubiquitous messaging platform — and a website, although Weng told TechCrunch in an interview this week that the company is contemplating a standalone app of its own. The benefit of that, beyond a potentially more engaging customer experience, could be to broaden LemoonBox’s product selection and use data to offer a more customized selection of products. Related to that, LemonBox said it hopes to work with health and fitness-related services in the future to gather data, with permission, to help refine the personal approach.

LemonBox’s team has now grown to 20 people, with 12 full-time staff and 8 interns, and Weng said that the new funding will also go towards increased marketing, improvements to the WeChat app and upgrading the company’s supply chain. Business, he added, is growing at 35 percent per week as LemonBox has adopted a personal approach to its packaging, much like Amazon-owned PillPack.

“This is the first time people in China have ever seen this level of customization for their vitamins,” Weng told TechCrunch.

Members of the LemonBox team with Qi Lu, who heads up Y Combinator’s China business

Qi Lu, the former Microsoft and Baidu executive who leads YC’s new China unit, said he is “bullish” about the business.

“What LemonBox offers resonates with me and is serving a clear China market needs. Personally, I travel a lot between China and the U.S, and I often was asked by my relatives to help purchase and carry them similar products like vitamins,” he said in a prepared statement.

“More importantly, what LemonBox can do is to build an initial core user base and a growing brand. Over time, by serving their users well, it can reach and engage more users who want to better take care of their broader nutrition needs, use more data and take advantage of increasingly stronger AI technologies to customers and personalize, and become an essential service for more and more users and customers in China,” Lu added.

A slew of venture capitalists known for high-profile exits — Kirsten Green of Forerunner Ventures, Keith Rabois of Khosla Ventures, Alfred Lin of Sequoia Capital and Alex Taussig of Lightspeed Venture Partners — have invested in Faire (formerly known as Indigo Fair), a 2-year-old wholesale marketplace for artisanal products.

A quick glance at Faire suggests it’s a combination of Pinterest and Etsy, complete with trendy, pastel stationery, soap, baby products and more, all made by independent artisans and sold to retailers. Faire has today announced a $100 million fundraise across two financing rounds: a $40 million Series B led by Taussig at Lightspeed and a $60 million Series C led by Y Combinator’s Continuity fund. New investors Founders Fund, the venture firm founded by Peter Thiel, and DST Global also participated. The business has previously brought in a total of $16 million.

The latest financing values Faire at $535 million, according to a source familiar with the deal.

If you’re feeling a little bit of déjà vu, that’s because a similar startup also raised a sizeable round of venture capital funding, announced today. That’s Minted . The 10-year-old company, best known for its wide assortment of wedding invitations and stationery, raised $208 million led by Permira, with participation from T. Rowe Price. Though Minted is first and foremost a consumer-facing marketplace, it plans to double down on its wholesale business with its latest infusion of capital, setting it up to be among Faire’s biggest competitors.

Like Minted, Faire leverages artificial intelligence and predictive analytics to forecast which products will fly off its virtual shelves in order to to source and manage inventory as efficiently as possible. The approach appears to be working; Faire says it has 15,000 retailers actively purchasing from its platform, including Walgreens, Walmart, Sephora and Nordstrom — a 3,140 percent year-over-year increase. It’s completed 2,000 orders to date, garnering $100 million in run rate sales, and has expanded its community of artists 445 percent YoY, to 2,000.

The company, headquartered in San Francisco, with offices in Ontario and Waterloo, was founded by three former Square employees: chief executive officer Max Rhodes, who was product manager on a variety of strategic initiatives, including Square Capital and Square Cash; chief information officer Daniele Perito, who led risk and security for Square Cash; and chief technology officer Marcelo Cortes, a former engineering lead for Square Cash.

“Our mission at Faire is to empower entrepreneurs to chase their dreams,” Rhodes wrote in a blog post this morning. “We believe entrepreneurship is a calling. Starting a business provides a level of autonomy and fulfillment that’s become difficult to find for many elsewhere in the economy. With this in mind, we built Faire to help entrepreneurs on both sides of our marketplace succeed.”

Vector’s already got the adorable bit down, courtesy of a team staffed by former Pixar and Dreamworks animators. But Anki’s grownup version of Cozmo can use a bit of help when it comes to smarts. Starting next week, the desktop robot will be able to tap into Amazon Alexa’s deep base of answers.

The new feature will arrive as a software update on December 17. As evidenced by the below preview video, it’s not a direct integration into Vector’s AI personality. “Hey Vector” will get you an answer from the familiar robot face, while “Alexa” will pop up a blue ring.

From there, you can do the standard array of Alexa features, from setting reminders to controlling smart home devices. Though, as the company notes, a handful of functions won’t be available at launch, including music streaming, Drop In, Calling and Kindle and Audible features.

Just in time for the holidays, the new additions should still make Vector a bit more useful day to day, expanding functionality beyond an adorable rolling robotic paper weight.

The company behind the biggest screens in cinema is giving up on bringing VR screens within a few inches of users’ faces. The company announced today in a SEC filing that it will be shutting down its three remaining virtual reality centers including its flagship location in Los Angeles.

Via the filing:

In connection with the Company’s previously-announced strategic review of its virtual reality pilot initiative, the Company has decided to close its remaining VR locations and write-off certain VR content investments.

The locations in LA, Bangkok and Toronto will be shuttered in Q1 of 2019 according to Variety.

After making a lot of noise about the centers at launch, the company seemed to realize pretty quickly that the economics just weren’t there. Previous to today’s announcement, IMAX had already shut down 4 of the 7 VR centers that had been opened.

A lot of virtual reality startups who were counting on the pipe dream resurgence of the American arcade scene are probably sweating a bit after today’s news. It was clear that IMAX’s efforts hadn’t been a raving success, but there’s a big difference between dialing it back and shutting it down.

Earlier this year, IMAX confirmed that it had paused work on a VR camera project it was developing with Google.

Machine learning models are getting quite good at generating realistic human faces — so good that I may never trust a machine, or human, to be real ever again. The new approach, from researchers at Nvidia, leapfrogs others by separating levels of detail in the faces and allowing them to be tweaked separately. The results are eerily realistic.

The paper, published on preprint repository Arxiv (PDF), describes a new architecture for generating and blending images, particularly human faces, that “leads to better interpolation properties, and also better disentangles the latent factors of variation.”

What that means, basically, is that the system is more aware of meaningful variation between images, and at a variety of scales to boot. The researchers’ older system might, for example, produce two “distinct” faces that were mostly the same except the ears of one are erased and the shirt is a different color. That’s not really distinctiveness — but the system doesn’t know that those are not important pieces of the image to focus on.

It’s inspired by what’s called style transfer, in which the important stylistic aspects of, say, a painting, are extracted and applied to the creation of another image, which (if all goes well) ends up having a similar look. In this case, the “style” isn’t so much the brush strokes or color space, but the composition of the image (centered, looking left or right, etc) and the physical characteristics of the face (skin tone, freckles, hair).

These features can have different scales, as well — at the fine side, it’s things like individual facial features; in the middle, it’s the general composition of the shot; at the largest scale, it’s things like overall coloration. Allowing the system to adjust all of them changes the whole image, while only adjusting a few might just change the color of someone’s hair, or just the presence of freckles or facial hair.

In the image at the top, notice how completely the faces change, yet obvious markers of both the “source” and “style” are obviously present, for instance the blue shirts in the bottom row. In other cases things are made up out of whole cloth, like the kimono the kid in the very center seems to be wearing. Where’d that come from? Note that all this is totally variable, not just a A + B = C, but with all aspects of A and B present or absent depending on how the settings are tweaked.

None of these are real people. But I wouldn’t look twice at most of these images if they were someone’s profile picture or the like. It’s kind of scary to think that we now have basically a face generator that can spit out perfectly normal looking humans all day long. Here are a few dozen:

It’s not perfect, but it works. And not just for people. Cars, cats, landscapes — all this stuff more or less fits the same paradigm of small, medium, and large features that can be isolated and reproduced individually. An infinite cat generator sounds like a lot more fun to me, personally.

The researchers also have published a new dataset of face data: 70,000 images of faces collected (with permission) from Flickr, aligned and cropped. They used Mechanical Turk to weed out statues, paintings, and other outliers. Given the standard dataset used by these types of projects is mostly red carpet photos of celebrities, this should provide a much more variable set of faces to work with. The dataset will be available for others to download here soon.