Year: 2018

Today at the Google Next conference in San Francisco, the company announced it would soon be enhancing Google meeting hardware to allow voice commands.

For many people setting up meetings remains a major problem and pain point. The company wants to bring the same voice-enabled artificial intelligence it uses for tools like Google Assistant to meeting hardware. To that end, the company introduced Voice Commands for Meet today.

This will allow users to say, “Hey Google, start the meeting.” And this is just a starting point. They promise to be adding additional commands over time. They will be adding this functionality later this year.

Just last Fall, the company launched the Hangouts Meet hardware program, which provided a way for Meet customers to launch meetings using Google or other hardware such as the traditional Cisco or Polycom hardware found in many conference rooms. Google reports that customers have set up thousands of these Hangouts Meet-enabled meeting rooms.

By providing some simple commands to set up the meeting, invite participants, join a meeting and so forth using your voice, it can greatly simplify the sometimes complicated activity of meeting administration, which even after all this years often seems unnecessarily complicated and frustrating for many people.

Users are certainly getting used to interacting with devices thanks to Google Home, the Amazon Echo and similar devices.

It’s worth noting that Google is not alone in trying to bring voice-enabled hardware into the meeting room. Last November, Cisco announced Cisco Spark Assistant to bring voice commands specifically to Cisco meeting room hardware. The underlying the voice recognition technology comes from the MindMeld acquisition, a conversational AI startup that Cisco bought in May 2017 for $125 million.

Last month, Uber hit 10 billion completed trips across rides and deliveries, the company announced today. This comes about one year after Uber hit five billion trips. Lyft, Uber’s main competitor in the U.S., hit half a billion rides last October.

Uber officially hit the milestone on June 10, 2018 when 173 trips and deliveries simultaneously started at 10:12pm GMT. The trips were in 21 countries across five continents.

This milestone is despite Uber’s many struggles since it first launched in 2010. As many of you are probably aware, Uber has faced a bevy of criticism in recent history pertaining to its workplace culture, the behavior of its former CEO Travis Kalanick and more.

If you want to hear more about Uber from the boss himself, be sure to hit up TC Disrupt, where I’ll be chatting with Uber CEO Dara Khosrowshahi.

Apple came out swinging when it announced its latest MacBook Pro update earlier this month. In preparation for the launch, the company seeded souped up versions of the notebook with a diverse array of creative professionals, who happily discussed the ways in which the new specs have improved their workflows.

The reviews have been largely positive, as well — our own included. The laptop performed admirably and racked up some impressive benchmark scores in our testing that were in line with the performance boosts recorded by Geekbench itself. But in amongst those reviews, Dave “D2D” Lee reported some troubling findings in a video titled “Beware the Core i9.”

The YouTuber found that exporting video with Premier Pro actually took longer on the new models sporting a top of the line Core i9 than they did on last year’s model running an i7. Counterintuitive to say the least — and something that appeared to be some sort of system throttling, in order to deal with overheating while exporting. In the video, Lee speculated that the laptop’s chassis (unchanged from the previous generation) simply couldn’t cool the i9 properly.

“It was an issue that was unique to this year’s version, particularly the i9,” Lee told TechCrunch, in a followup interview this week. “It only occurred because this particular CPU is as power-hungry as it is. That puts it over the top.”

Benchmarks often only tell part of the story. They work well as for sterile, laboratory testing, but don’t always offer a decent stand-in for real world usage. Among other things, workflows vary a good deal from case to case and user to user. Lee says he was surprised to find that his own workflow was taxing to the system compared to the ones others shared in the online community after the initial video was posted.

Even so, as Apple makes a major push to recapture the hearts and minds of creatives, it will require systems that can handle just about anything creatives can throw at it. Lee tells TechCrunch that both Apple and Premier Pro-producer, Adobe, were quick to reach out after the video was posted.

“It was impressive to see how quickly they worked with the community find out what was going on,” he says. “The first step was to try to replicate this, to find out what was causing it. When they figured that out, I think they tried to figure out how to resolve it.”

It seems that Apple’s initial testing simple didn’t account for the specifics of the workflow Lee put the system through. In the intervening days, however, Apple says it has identified the issue and will be releasing a fix in an update to MacOS High Sierra rolling out today.

The company apologized for the bug fix in a statement provided to TechCrunch,

Following extensive performance testing under numerous workloads, we’ve identified that there is a missing digital key in the firmware that impacts the thermal management system and could drive clock speeds down under heavy thermal loads on the new MacBook Pro. A bug fix is included in today’s macOS High Sierra 10.13.6 Supplemental Update and is recommended. We apologize to any customer who has experienced less than optimal performance on their new systems. Customers can expect the new 15-inch MacBook Pro to be up to 70% faster, and the 13-inch MacBook Pro with Touch Bar to be up to 2X faster, as shown in the performance results on our website.

The company notes that a majority of users experience performance gains — which we can certainly confirm in our own testing. The issue at hand appears to have been something more complex and rarer than just a system shutting down or cutting back on power from overheating. The company is recommending users install the Supplemental Update to avoid performance slow downs during taxing actions.

The 24-hour clock is in play ladies, gentlemen and everyone in-between. That’s how long you have before our early-bird pricing flies south. If you want to go to Disrupt San Francisco 2018 on September 5-7, buy your tickets before July 25 at 5 p.m. PST. Depending on the type of pass you purchase, you can save up to $1,200. Miss that deadline and it’s full-freight for you. Why say “ouch” if you don’t have to? Get your ticket today.

This conference will be the only Disrupt event in North America this year, so we’ve gone all-out to make it memorable. We’ve relocated to Moscone Center West and tripled our floor space in the process. The space will be comfortable and ready for the 10,000 attendees — including more than 1,200 exhibitors and 400 media outlets — heading our way.

There are plenty of excellent reasons to attend Disrupt SF 2018. Come for the great speakers, like Ben Horowitz, co-founder of the venture firm Andreessen Horowitz, and author of “The Hard Thing About Hard Things,” or Priscilla Chan, of the Chan Zuckerberg Initiative. You’ll find dozens more in the full lineup of Disrupt SF 2018 speakers.

Come for Startup Battlefield, the premier startup competition and the best platform for launching early-stage startups to the world. This year, we’ve doubled the prize money, and one spectacular startup will walk away with $100,000 in non-equity cash. As you might imagine, our vetting process is competitive, to say the least. We can’t wait to tell you more about this cohort, but you’ll have to wait until we complete our evaluations. Stay tuned!

Come for the Virtual Hackathon. Yes, you heard right. We’ve gone virtual to go global. The Virtual Hackathon is open to anyone anywhere in the world. Right now, thousands of hackers, programmers, designers and developers are creating amazing products, and we can’t wait to see what they dream up.

Come to see those 1,200 exhibitors in Startup Alley, the pulsing heart of every Disrupt. And while you’re there, be sure to check out the 50+ startups that earned a TechCrunch Top Pick designation. You’ll find five Top Picks representing each one of these 12 categories: AI, AR/VR, Blockchain, Biotech, Fintech, Gaming, Healthtech, Privacy/Security, Space, Mobility, Retail or Robotics.

There’s so much to see and do, and not nearly enough time to tell you everything. Not to worry, you can always check out the full Disrupt San Francisco 2018 agenda.

Disrupt San Francisco 2018 takes place on September 5-7 at Moscone Center West. The 24-hour clock is ticking, and the prices go up on July 25 at 5 p.m. PST. Don’t miss your chance to get the best price possible. Buy your tickets today.

Josh Mohrer (pictured above) and William Barnes, two former Uber executives, are working on an investor syndicate to invest in startups led by fellow former Uber employees, Axios reported and TechCrunch has confirmed.

“We believe that people who help build transformative companies will go on to do other awesome things,” Mohrer told TechCrunch about the syndicate’s thesis to invest in Uber alumni. “The idea is that we can really move the needle for companies who are doing things that are Uber-adjacent — things that are in our wheelhouse.”

Mohrer and Barnes have been working on this for the last nine months or so. While a lot of the emphasis is on backing startups led by former Uber employees, that’s not a strict requirement, Mohrer said.

The syndicate, which is made up of around 100 former Uber employees, “started pretty organically,” Mohrer said. So far, about 50 people have taken part in at least one investment. The check sizes have been modest — in the hundreds of thousands of dollars.

The focus of the syndicate is on two-sided marketplaces and transportation startups like Lime and Cargo, which partnered with Uber last week to enable drivers to sell passengers goods during rides. The syndicate’s other investments are in Replicated, Service and Salido.

“In the fall, we’re strongly considering what a traditional VC angle on this would look like,” Mohrer said. Likely, that would be called Moving Capital.

Last May, Mohrer left Uber to join Tusk Ventures as its managing director. Before joining Tusk, Mohrer spent about five years at Uber. While at Tusk, Mohrer led the firm’s Series A and B investments in Lime competitor Bird. Barnes, on the other hand, formerly led Uber’s West Coast operations for almost six years.

Cloud Functions, Google’s serverless platform that competes directly with tools like AWS Lambda and Azure Functions from Microsoft, is now generally available, the company announced at its Cloud Next conference in San Francisco today.

Google first announced Cloud Functions back in 2016, so this has been a long beta. Overall, it also always seemed as if Google wasn’t quite putting the same resources behind its serverless play when compared to its major competitors. AWS, for example, is placing a major bet on serverless, as is Microsoft. And there are also plenty of startups in this space, too.

Like all Google products that come out of beta, Cloud Functions is now backed by an SLA and the company also today announced that the service now runs in more regions in the U.S. and Europe.

In addition to these hosted options, Google also today announced its new Cloud Services platform for enterprises that want to run hybrid clouds. While this doesn’t include a self-hosted Cloud Functions option, Google is betting on Kubernetes as the foundation for businesses that want to run serverless applications (and yes, I hate the term ‘serverless,’ too) in their own data centers.

It used to be that developers built applications with long lead times and development cycles. There was always plenty of time to prepare, but in today’s continuous delivery/continuous deployment (CI/CD) world, new versions could be going out every day. That requires a CI/CD framework, and today at Google Next in San Francisco, the company announced Cloud Build, its new CI/CD framework.

As Google describes it, Cloud Build is the company’s “fully-managed Continuous Integration/Continuous Delivery (CI/CD) platform that lets you build, test, and deploy software quickly, at scale.”

Cloud Build works across a variety of environments including VMs, serverless, Kubernetes, or Firebase. What’s more it supports Docker containers and it gives developers or operations the flexibility to build, test and deploy in an increasingly automated fashion.

Google will allow you to use triggers to deploy, so that when certain conditions are met, the update will launch automatically. You can identify vulnerabilities in your packages before you deploy and you can build locally and deploy in the cloud if you so choose.

If there are problems, Cloud Build provides analytics and insights to let you debug via build errors and warnings and filter those warnings to easily identify slow builds or those with other issues you want to see before deploying live.

Google is offering a free version of Cloud Build with up to 120 build minutes per day at no cost. Additional build minutes will be billed at $0.0034 per minute.

As Google pushes further and further into enterprise services, it’s looking to leverage what it’s known for — a strong expertise in machine learning — to power some of the most common enterprise functions, including contact centers.

Now Google is applying a lot of those learnings in a bunch of new updates for its contact center tools. That’s basically leaning on a key focus Google has, which is using machine learning for natural language recognition and image recognition. Those tools have natural applications in enterprises, especially those looking to spin up the kinds of tools that larger companies have with complex customer service requests and niche tools. Today’s updates, announced at the Google Cloud Next conference, include a suite of AI tools for its Google Cloud Contact Center.

Today the company said it is releasing a couple of updates to its Dialogflow tools, including a new one called phone gateway, which helps companies automatically assign a working phone number to a virtual agent. The company says you can begin taking those calls in “less than a minute” without infrastructure, with the rest of the machine learning-powered functions like speech recognition and natural language understanding managed by Google.

Google is adding AI-powered tools to the contact center with agent assistant tools, which can quickly pull in with relevant information, like suggested articles. It also has an update to its analytics tools, which lets companies sift through historical audio data to pull in trends — like common calls and complaints. One application here would be to be able to spot some confusing update or a broken tool based on a high volume of complaints, and that helps companies get a handle on what’s happening without a ton of overhead.

Other new tools include sentiment analysis, correcting spelling mistakes, tools to understand unstructured documents within a company like knowledge base articles — streaming that into Dialogflow. Dialogflow is also getting native audio response.

It pays to be a monopolist.

Alphabet’s earnings were stellar, and that is truly saying something. Just a few weeks ago, the European Union placed a record €4.34 billion fine on the Mountain View-based company, a penalty for the company’s payments to OEMs to include Google Search as the default search option in order to access Google Play, the company’s App Store.

The acrimonious feud with the EU has become such a constant financial concern for the company that it now includes a “European Commission fines” line item in its consolidated statements of income.

Yet, one can’t help but stand back in awe at a company whose results show the complete lack of teeth of existing antitrust law, whether here in America or anywhere else globally. Alphabet’s revenues grew by $6.6 billion, far more than the record fine the EU laid on the company. Net income for the quarter was $3.2 billion even after the fine was deducted as an expense. The Alphabet cash machine remains as strong as ever.

The EU fine was of course one component in the plan of the antitrust authorities. There are structural remedies, namely that Alphabet needs to cease and desist on leveraging Android to cement its market share in search. But at this point, what exactly are the alternatives for handset manufacturers? DuckDuckGo? Bing?

My colleague Natasha Lomas along with many other journalists discussed the potential of the EU demanding that Alphabet being broken up. Yet, even such a meat cleaver of a structural remedy would seem to be useless at this juncture. Google Search essentially has no peer, and isn’t likely to have one in the near future. It has brand equity, data equity, extensive capital investments, and trade secrets. No amount of structural remedies save the complete destruction of the company is going to reduce those burdens to competition.

These fines then are less about punishing behavior — after all, they aren’t deterring would-be monopolists from their activities. Instead, they essentially act as an excess profits tax, a way to uniquely target extraordinarily profitable tech companies without changing general business taxes.

Even when we expand the lens beyond just these anticompetitive enforcement actions to include data sovereignty issues like GDPR, Alphabet is once again positioned to be a winner. As I have written before, Alphabet and Amazon are likely the only companies with sufficient scale to even begin to handle the myriad laws and regulations emanating around the world on data sovereignty. Far from empowering consumers, these laws essentially ensure that there is now an added “regulatory network effects” barrier to competition in these markets.

The next billion internet users will ultimately determine the ceiling for Alphabet’s revenues

To me, there is only one force today that has any potential to threaten Alphabet’s complete and on-going dominance, and that is China and its ambitious tech industry. Transsion’s subsidiaries dominate in the African smartphone market, and it along with other smartphone players like Xiaomi have targeted India and its quickly burgeoning middle class. If the next one to two billion internet users come to rely on Chinese internet services instead of Alphabet, that could prove a serious competitive headwind for the company.

One legacy of GDPR may simply be that it forced large tech companies to double down on the U.S. and Europe at a time when they should have been focused on global expansion. Alphabet broke the $5 billion revenue barrier for the Asia-Pacific region for the first time this quarter, but that amounts to only 15.6% of the company’s revenues. Meanwhile, Facebook, dealing with its Cambridge Analytica imbroglio, has started to curtail the expansion of its Free Basics internet access scheme.

Those distractions provide a rare opportunity for Chinese companies to focus exclusively on global expansion. Certainly Huawei and ZTE have taken that course. While broadly blocked from the U.S. market and with Australia preparing to ban 5G deployments, the two have had tremendous success in developing markets, with infrastructure and handset products that are often significantly cheaper than competitors.

All of that might mean little to the U.S. or European consumer, but it does potentially put a ceiling on the growth of Alphabet and other large tech companies. As TechCrunch pointed out yesterday, there has been a race to see who will break the trillion dollar market cap barrier first among the major tech players. Alphabet is sitting at $865 billion and a trillion isn’t far away. But could it grow much beyond that? That to me depends on these new, developing markets, and there the race is much more competitive.

As these earnings show, the jaws of antitrust have no teeth, and competitive dynamics might constrain Alphabet to merely be a trillion dollar company. It pays — over and over again — to be a monopolist.

Twitch wants more people to stream, so it’s going to begin teaching them how. The video game streaming site today announced the launch of Twitch Creator Camp, a new educational resource that helps newcomers learn the basics of streaming, as well as how to build up a channel, connect with fans, and earn rewards.

The launch of the how-to site comes about a week after an article by The Verge detailed the long tail of Twitch streamers, with a focus on those who spend years broadcasting to no one in the hopes of one day gaining a following.

The article raised the question that, in the age of live streaming, where every major social company – including Facebook, Instagram and YouTube – today offers easy streaming tools, there many not be enough of an audience for all the content creators are producing.

Twitch, apparently, believes the issue is one that can be addressed – at least in part – by training new streamers.

On Twitch Creator Camp, the company is bringing in successful creators to help educate the would-be streamers on a variety of often-discussed topics. These insights will be shared as articles, videos and live streams.

At launch, the site includes content focused on a variety of streaming best practices, including the basics of setting up a channel, building a brand, leveraging their stats, using Twitch features like emotes, badges and extensions, and more.

Streamers will also learn how to better network with others and engage their audience, as well as how to optimize their channel for monetization through subscriptions, merchandise, ads and sponsorships.

In addition, creators will begin live streaming on Creator Camp, starting on July 31 at 2 PM PT.

At this time, a number of Twitch Partners will answer general questions about streaming. A calendar of upcoming streams is also available on Twitch’s site, as the company aims to host weekly sessions going forward.

“Hosting a good stream isn’t easy. We’ve heard from many of our creators that they spend a lot of time searching for advice on effective tools, features, and techniques in order to make their broadcasts more engaging and to grow their communities,” said Jessica Messinger, Creator Growth Marketing Manager at Twitch, in a statement.

“Twitch Creator Camp makes things simpler by centralizing the most relevant information to a creator’s success, all of which is provided by Twitch and many of our successful Partners. We want to help our creators succeed and this is just the beginning,” she added.

Twitch says the partners it’s working with for Creator Camp are being compensated for their efforts. Currently, those participating include: Jericho, gassymexican, teawrex, JGhosty, pokket, firedragon, venalis, tominationtime, sypherpk, xmiramira, iamBrandon, DeejayKnight, Lobosjr, sacriel, PmsProxy, itmeJP, kaypealol, and Pokimane.

Twitch today has over 2.2 million broadcasters serving up streams on its site every month, which are consumed by 15 million daily active viewers who watch an average of 95 minutes of content daily. However, much of the on-site activity – just like on YouTube and elsewhere – is dominated by top creators.

Meanwhile, many of Twitch’s smaller streamers may already understand the basics and tips that Twitch’s Creator Camp is offering. For them, the issue is not one of following all the steps being laid out, but rather one of discovery.

Twitch has been working to address its discovery issues, too, having last month detailed a number of projects it’s working on across this front which are in various phases of development.

“We don’t believe Twitch should be a popularity contest” the company said at the time.

Twitch Creator Camp is open as of today, with the live streams starting at the end of the month.