Category: UNCATEGORIZED

Atlassian has a portfolio of developer tools like Bitbucket, Jira and Confluence. It also has a marketplace with hundreds of add-ons, but what it lacked was a development platform to call its own. Today, that changed when the company announced the Forge platform.

“Forge will empower developers to more easily build and run enterprise-ready cloud apps that integrate with Atlassian products,” the company wrote in a blog post announcing the new tools.

The platform consists of three main components. For starters, it’s providing a serverless Function as a Service (FaaS) for developers to build hosted applications on Forge without worrying about the underlying infrastructure resources required to run the applications. The tool is actually built on AWS Lambda, AWS’s FaaS.

This should allow more developers to get involved because it strips away a layer of complexity around managing infrastructure. “A FaaS platform also lets us eliminate common pain points such as authentication, identity, scaling and tenancy,” the company wrote in the blog post.

The tool kit also includes a UI component called Forge UI for building user interfaces on the web or devices. Forge UI uses a declarative language that should make it easier to build user interfaces, and as with the function layer, the idea here is to simplify the process for users. Atlassian will deal with all of the security involved in building a user interface, something that many developers struggle with.

“By abstracting away the process of rendering the UI layer, Forge makes stronger guarantees about how apps present or transmit sensitive data, such as user-generated content and personally identifying information,” the company wrote.

The final piece is a command line interface (CLI) called Forge CLI. The idea here is to build continuous delivery pipelines with Bitbucket and run them from the command line. If you put all three of these components together, you have a pretty comprehensive development environment with tools for building functionality and designing user interfaces, while managing operations from a command line.

There are lots of platform service offerings out there, so Atlassian faces some competition here, but for developers who planned on building apps for the Atlassian marketplace, this set of tools could prove useful and help push more developers to join in.

With just under a year until U.S. Election Day, Twitter is bringing back its Election Labels, which provide information about political candidates — including what office they’re running for and their state and district number. The labels will also have a small ballot box icon to accompany this information. The feature was first launched during the 2018 U.S. midterms, where the labels were seen 100 million times per day by Twitter users in the week before Election Day.

In addition, 13% of U.S. election-related conversations on Twitter included a tweet with an Election Label, the company says.

Now the labels are making a return ahead of the 2020 U.S. elections.

The labels will appear on accounts of candidates who are running for the U.S. House of Representatives, U.S. Senate, or Governor in the 2020 election who have qualified for the general election ballot, says Twitter. And they will begin to appear on candidates’ Twitter accounts after they qualify, which will happen on a rolling basis as states have different caucus and election dates, the company notes. The first takes place on March 3rd.

To enable the feature, Twitter has again partnered with Ballotpedia, a civic nonprofit that publishes non-partisan information about federal, state and local politics. The organization will help Twitter to identify which candidates have qualified for the general election ballot so their accounts can be appropriately labeled.

The Election Label will appear on the profile page of a candidate’s Twitter account and on every tweet and retweet they post to their account, even when embedded on sites off of Twitter.

Ahead of this, Twitter today will also start to verify the campaign Twitter accounts of those who have qualified for primary elections for the U.S. House, Senate or Governor. This is different from how Twitter handled candidate verification during the 2018 midterms. Back then, it only verified candidates after they qualified for the general election ballot. This time around, Twitter says it will proactively verify the primary candidates.

This verification is the same checkmark other high-profile accounts receive — like those belonging to celebrities or other public figures. These verifications will start today and will continue on a rolling basis as states have different filing deadlines. Ballotpedia is also assisting on this effort as well, by helping Twitter identify the candidates.

Twitter, like other social platforms, had been heavily impacted by foreign interference with the U.S. 2016 presidential election. Last year, Twitter said that 1.4 million people had interacted with Russian trolls during the presidential campaign, which is more than double the 677,775 that Twitter originally believed had either seen, followed, or retweeted one of those accounts. These interference issues have been ongoing, as thousands of Twitter accounts spreading false information remained active in the weeks ahead of the U.S. midterms.

Bots continue today to infect the platform, in an effort to sway public opinion. For example, in April, Twitter removed over 5,000 bots with ties to a social media operation that previously promoted messages sympathetic to Saudi Arabia’s government. The bots had more recently been promoting the “Russiagate” hoax.

Disinformation efforts like this are not just impacting social platforms in the U.S. nor are they only associated with Russian bots. In a report released at the beginning of 2019, Twitter said it had banned more than 4,000 disinformation accounts originating in Russia, 3,300 from Iran, and more than 750 from Venezuela.

When Twitter first introduced the Election Labels for the U.S. midterms, it stressed how important it is for people using its platform to be able to identify the original sources and authentic information.

Today, Twitter’s system to label and verify politicians and candidates’ campaigns is now a part of a number of efforts Twitter has underway to make sure conversations taking place on its platform are authentic. The company says it will later release more tools to help better find quality news and have more informative conversations on Twitter.

Launch provider Rocket Lab has opened the doors on LC-2, its first launch facility in the U.S., adding capacity and versatility for providing trips to orbit. And LC-2 already has its first customer: the U.S. Air Force’s Space Test Program.

The company had a little shindig today at the facility, located on Wallops Island in Virginia — home to NASA’s Wallops Flight Facility as well. There they took the wrapper off LC-2, which has been under construction since it was announced last October.

The team breaks ground back in 2018.

It’s not some wild new concept, just a typical launchpad and support facilities where the rockets live, get checked, fueled, and so on. The most important difference with this one, for Rocket Lab, is that it’s here in the U.S.; So far, all its 10 commercial launches have been from Launch Complex 1 in New Zealand, where the company is based.

The new facility will be put to use soon: the Air Force is first in line to put a payload into orbit in a launch currently planned for Q2 of 2020. All we know about the mission, STP-27RM, is that it will “test new capabilities that we will need in the future.”

“It’s an honor and privilege to be launching a U.S. Air Force’s Space Test Program payload as the inaugural mission from Launch Complex 2,” said Rocket Lab founder and CEO Peter Beck in a press release. “We’ve already successfully delivered STP payloads on Electron from Launch Complex 1, and we’re proud to be providing that same rapid, responsive, and tailored access to orbit from U.S. soil.”

Right now LC-2 is “only” equipped to handle up to 12 launches a year, while LC-1 can theoretically do 120. Rocket Lab is nowhere near hitting those rates just yet, but they’re well on their way with a perfect track record and a demonstration of its ability to quickly adjust timeframes. The goal is eventually to be launching weekly, or even more frequently than that. And the more launch sites they have to pull that off, the better.

Amazon is having another go at expanding its reach to listeners in India. The company, which launched pay-to-use Audible in the country last year, today introduced a new service called Audible Suno that offers free access to “hundreds of hours of audio entertainment, enlightenment and learning.”

And it’s banking on major Indian celebrities to draw the listeners.

Audible Suno, which is exclusively available to users in India, features more than 60 original and exclusive series in both Hindi and English languages. Audible, the world’s largest seller and producer of audio content, said Suno is aimed at filling the “idle time” listeners have each day during their commutes and performing other daily chores.

The company says Audible Suno is also free of advertisements.

The launch of Audible Suno in India illustrates the commitment the company has in the country, said Audible founder and CEO Don Katz. Amazon has invested more than $5.5 billion in its business in India to date. The company’s tentacles today reach a number of categories in the country including e-commerce, payments, online ticketing business, video and audio streaming, and VC deals.

“I’ve always been passionate about the transformative power of the spoken word, and I’m delighted to be able to offer this breadth of famous voices and culturally resonant genres with unlimited access, ad-free and free of charge,” Katz said in a statement.

Who are these famous voices you ask? Here’s the list: Amitabh Bachchan (Kaali Awaazein), Katrina Kaif, Karan Johar (Picture Ke Peechhe), Anil Kapoor, Farhan Akhtar, Mouni Roy, Anurag Kashyap, Tabu, Nawazuddin Siddiqui, Diljit Dosanjh, Vir Das and Vicky Kaushal.

Audible Suno is available to users through a dedicated Android app and via iOS Audible app.

More to follow…

Today at TechCrunch Disrupt Berlin, Samsung Electronic’s President Young Sohn revealed the company had sold 1 million foldable Galaxy Fold smartphones. Estimates from October pegged sales at that time at 500,000 units.

“And I think that the point is, we’re selling [a] million of these products,” Sohn said. “There’s a million people that want to use this product at $2,000.”

Today’s conversation at Disrupt Berlin focused around growth through innovation. Sohn commented on the sales number while explaining Samsung’s process of releasing products to get feedback. He said, in part, if they kept devices like the Fold in labs, they wouldn’t get the input they needed.

And Samsung got a lot of feedback about the Galaxy Fold.

The foldable phone was first announced early this year at MWC 2019, where it was among a handful of foldable devices. It launched several months later in April, where reviewers quickly discovered multiple problems, including screens that cracked. The company soon (though perhaps not quickly enough) reworked the product, rereleasing it in late September.

The rereleased Galaxy Fold was more durable, though our review unit still had screen issues.

Today at Disrupt Berlin, I asked if Samsung is comfortable selling a $2,000 device that is essentially a beta device. He said yes, and pointed to the new sales number as justification.

Previous media reports stated Samsung is ramping up plans to sell 6 million foldable devices in 2020.

At the very beginning, there were 14 startups. After two days of incredibly fierce competition, we now have a winner.

Startups participating in the Startup Battlefield have all been hand-picked to participate in our highly competitive startup competition. They all presented in front of multiple groups of VCs and tech leaders serving as judges for a chance to win $50,000 and the coveted Disrupt Cup.

After hours of deliberations, TechCrunch editors pored over the judges’ notes and narrowed the list down to five finalists: Gmelius, Hawa Dawa, Inovat, Scaled Robotics and Stable.

These startups made their way to the finale to demo in front of our final panel of judges, which included: Andrei Brasoveanu (Accel), Andrew Reed (Sequoia Capital), Carolina Brochado (SoftBank Vision Fund), Lila Preston (Generation Investment Management) and Mike Butcher (TechCrunch).

Winner: Scaled Robotics

Scaled Robotics has designed a robot that can produce 3D progress maps of construction sites in minutes, precise enough to detect that a beam is just a centimeter or two off. Supervisors can then use the software to check things like which pieces are in place on which floor, whether they have been placed within the required tolerances or if there are safety issues like too much detritus on the ground in work areas.

Read more about Scaled Robotics in our separate post.

Runner-Up: Stable

Stable offers a solution as simple as car insurance, designed to protect farmers around the world from pricing volatility. Through the startup, food buyers ranging from owners of a small smoothie shop to Coca-Cola employees can insure thousands of agricultural commodities, as well as packaging and energy products.

Read more about Stable in our separate post.

The global industry potential of artificial intelligence is well-documented, yet the vision of this AI future is uncertain.

AI and automation trends are generating significant debate among economists and governments, particularly around employment impact and uncertain social outcomes. The mainstream attention is warranted. According to PwC, AI “could contribute up to $15.7 trillion to the global economy in 2030, more than the current output of China and India combined.”

AI is at a crossroads, and its long-term outlook is still hotly debated. Despite social media giants, automotive companies and numerous other industries investing hundreds of billions of dollars in AI, many automation technologies are not yet directly generating revenue and instead are forecasted to become profitable in the coming decades. This creates additional uncertainty of AI’s true market potential. The realistic potential value of AI is unknown, yet, as the technology advances, the ultimate impact could be of great consequence to virtually every economy.

There are many reasons to view AI’s future from an optimistic lens, however: chatbots provide significant evidence for AI’s positive impact on both business growth and employment markets. Today, chatbots are increasingly capable of mimicking human interactions and conversations to assist business-to-business, business-to-consumer, business-to-government, advertising audiences and other diverse groups. The evolution of the cognitive computer science behind conversational chatbots is perhaps one of the best examples of AI technologies driving revenue. Further, chatbot technology shows some of the greatest promise for augmenting, rather than replacing human workers.

AI is driving value while augmenting human workers

Chatbots are delivering real revenue today for some of the world’s leading financial services (Bank of America), retail (Levi’s), and technology companies (Zendesk) . We’re seeing more consumers taking the next step in a transaction or even making a purchase decision based off conversations with chatbots. Beyond driving sales, chatbots have numerous applications to a wide range of organizations. Nonprofits, NGOs, and even political campaigns find value in deploying chatbots to help handle the influx of inquiries from stakeholders and relevant audiences.

Rather than these chatbots replacing human workers, organizations are finding chatbots to be a helpful and value-creating opportunity that frees employees to focus on more strategic tasks. Apple’s Siri, Amazon Alexa and Microsoft Cortana aren’t replacing executive assistants today, but these technologies are all capable of supporting the executive assistant function in the workplace.

Gartner predicts AI augmentation, defined as a “human-centered partnership model of people and AI working together to enhance cognitive performance,” could generate $2.9 trillion of business value by 2021. Many industries see potential for chatbots to augment functions like sales, customer support and IT, enabling workers to create value in more strategic ways. Bain & Company finds chatbots to be among the most notable examples of artificial intelligence and automation in practice: “Companies use AI applications to understand industry trends, manage their workforce, address problems, power chatbots and personalize content to enable self-service.”

Clearly, the implications of scaled, human-like engagement are stunning in their capacity to carry out tasks. A chatbot’s ability to simultaneously hold tens of thousands of conversations — pulling from many millions of data points — is comparable to what a human customer service rep could accomplish in more than 1,000 years of nonstop work. Scaling customer service via AI allows service professionals to focus on big picture and more complex issues, and it provides rich data on customer interactions. We anticipate seeing more companies look to build better customer service experiences through chatbots, as Google and Salesforce announced in April.

The transformative impact of chatbots across industries

From our research and work with leading global companies, it’s clear that enterprises are finding that chatbots bring about tremendous value while supporting both people employment and long-term business growth opportunities today. Ultimately, chatbots are on track to showcase some of the most optimistic examples of AI augmentation. Consider three examples:

Hello and welcome back to our regular morning look at private companies, public markets and the grey space in between.

This morning we’re digging into the current IPO market, asking ourselves how much damage WeWork really did to other companies hoping to go public. Is the IPO window closed, and if not, what sort of companies can still get out?

There’s some good news out today for late-stage startups looking to debut — along with a few impending tests regarding the market’s appetite for risk that we should understand as we head into 2020.

Bill.com’s good news

In terms of IPOs, Bill.com’s felt comfortably standard for 2019. Bill.com was a heavily venture-backed company that had raised just under $350 million while private across myriad rounds, and by the time it wanted to go public it still lost money.

At the same time, the company had a number of strengths. These include historically slim losses as a percent of revenue ($7.3 million in its most recent fiscal year, against $78.4 million in revenue), differentiated revenue sources (subscription income and rising interest payments), and improving gross margins (74 percent in its most recent quarter, up from a little under 72 percent in the year-ago period).

Those factors combined were sufficient to entice investors to price the company’s IPO far above its initial expectations of $16 to $18 per share. Instead, Bill.com raised its range once and then priced above the higher interval. At $22 per share, the company’s value rose by about 60% compared to its most recent private financing. (You can read more on the debut here.)

This matters as WeWork was said to have closed the IPO window for companies more focused on growth than profits. The way the market reality was discussed in venture circles seemed to indicate that WeWork’s implosion had slashed investor interest in growth, with public market players now favoring profits, or something close.

Bill.com’s most recent three-month period featured far-larger losses than its year-ago quarter, which mattered little in the end. The firm’s solid growth and moderate losses, it seems, were more than enough to secure a strong welcome to the public markets.

Yes, but…

You may be wondering why we just spent so much time explaining why a healthy company managed to go public. The goal, simply, was to point out that not only can companies still losing money and burning cash go public, they may even get a strong reception.

But what about companies in slightly less good shape? What does Bill.com’s IPO pricing indicate for Sprout Social, a company of similar size that’s going public this week which is also unprofitable, but growing more slowly (29.5% year-over-year in Q3 2019, compared to Bill.com’s 57%)?

Its pricing and debut will be a more interesting test. And luckily for us, it should price its shares this evening. (Even more fun, it targeted the same $16 to $18 per-share initial IPO price range that Bill.com initially had in its own sights.)

If Sprout Social manages to price in-range, we’ll have another data point in favor of the IPO window being comfortably open. It’s not surprising that Bill.com’s IPO priced well, but Sprout Social’s slower growth rate likely make its losses less palatable; if it can debut all the same we’ll know that the band of venture-backed companies that can public post-WeWork in the dead of December is wide.

That’s good news for illiquid unicorns and their backers, provided that their companies are at least as healthy as Chicago’s Sprout.

WeWork 2.0

Finally, we have one more test of the IPO market ahead of us.

China-based Ucommune is a co-working company with self-described “global impact and ambitions.” Claiming to be the “largest co-working space community in China,” Ucommune espouses “sharing, innovation, responsibility and success for all.” In its F-1 document, filed yesterday and setting in motion a possible US-listed IPO, Ucommune details comical levels of unprofitability and growth.

If all that sounds familiar, it should. It should feel similar to WeWork, which makes the timing of Ucommune’s IPO filing all the more amazing. WeWork’s pulled IPO was minutes ago, and here we are, staring down the filing of yet another coworking IPO?

The situation gets even better. Observe the following results:

• Ucommune Q1, Q2, Q3 revenue: $122.4 million

Waymo has acquired Latent Logic, a UK company that spun out of Oxford University’s computer science department, as the autonomous vehicle company seeks to beef up its simulation technology.

The acquisition also marks the launch of Waymo’s first European engineering hub will be in Oxford, UK. This likely won’t be the end of Waymo’s expansion and investment in Europe and the UK. The former Google self-driving project that is now an Alphabet business said it will continue to look for opportunities to grow the team in the UK and Europe.

Earlier this year, Waymo locked in an exclusive partnership with Renault and Nissan to research how commercial autonomous vehicles might work for passengers and packages in France and Japan. In October, Waymo said that its working with Renault to study the possibility of establishing an autonomous transportation route in Paris.

Waymo has made simulation a one of the pillars of its autonomous vehicle development program. But Latent Logic could help Waymo make its simulation more realistic by using a form of machine learning called imitation learning.

Imitation learning models human behavior of motorists, cyclists and pedestrians. The idea is that by modeling the mistakes and imperfect driving of humans, the simulation will become more realistic and theoretically improve Waymo’s behavior prediction and planning.

Waymo isn’t sharing financial details of the acquistion. But it appears that the two founders Shimon Whiteson and João Messia, CEO Kirsty Lloyd-Jukes and key members of the engineering and technical team will join Waymo. The Latent Logic team will remain in Oxford.

After rolling out on smart speakers and displays earlier this year, Google’s interpreter real-time translation mode finally landing on mobile. A far more handy application for such functionality, the feature arrives on both Android and iOS handsets globally, starting today.

The feature works in tandem with Assistant. Say something like, “Hey Google, be my German translator” or “Hey Google, help me speak Thai,” and the feature kicks in, offering up a real-time translated transcript and audio. The feature also offers some Smart Replies a la Gmail, to help keep the conversation going.

The feature is now available in 44 languages (full list here), up from the 29 available on the smart displays/speakers. It’s integrated directly into the Google Assistant app, negating the need to download an additional translation app. Between this and Lens, Google’s apps have quickly become a necessary part of traveling abroad.