Tag Archives: FRANCISCO

Announcing new AI and mixed reality business applications for Microsoft Dynamics – The Official Microsoft Blog

Today, I had the opportunity to speak to press and analysts in San Francisco about our vision for business applications at Microsoft. In addition, I had the privilege to make two very important announcements: the upcoming availability of new Dynamics 365 AI applications, and our very first mixed reality business applications: Dynamics 365 Remote Assist and Dynamics 365 Layout.

Our vision for business applications at Microsoft

We live in a connected world where companies are challenged every day to innovate so they can stay ahead of emerging trends and repivot business models to take advantage of new opportunities to meet growing customer demands.

To innovate, organizations need to reimagine their processes. They need solutions that are modern, enabling new experiences for how they can engage their customers while making their people more productive. They need unified systems that break data silos, so they have a holistic view of their business, customers and employees. They need pervasive intelligence threaded throughout the platform, giving them the ability to reason over data, to predict trends and drive proactive intelligent action. And with adaptable applications, they can be nimble, allowing them to take advantage of the next opportunity that comes their way.

Two years ago, when we introduced Dynamics 365 we started a journey to tear down the traditional silos of customer relationship management (CRM) and enterprise resource planning (ERP). We set out to reimagine business applications as modern, unified, intelligent and adaptable solutions that are integrated with Office 365 and natively built on Microsoft Azure.

With the release of our new AI and mixed reality applications we are taking another step forward on our journey to help empower every organization on the planet to achieve more through the accelerant of business applications. Specifically, today we are making the following announcements:

Dynamics 365 + AI

First, I am happy to announce the coming availability of a new Dynamics 365 AI offering — a new class of AI applications that will deliver out-of-the-box insights by unifying data and infusing it with advanced intelligence to guide decisions and empower organizations to take informed actions. And because these insights are easily extensible through the power of Microsoft Power BI, Azure and the Common Data Service, organizations will be able to address even the most complex scenarios specific to their business.

Dynamics 365 AI for Sales: AI can help salespeople prioritize their time to focus on deals that matter most, provide answers to the most common questions regarding the performance of sales teams, offer a detailed analysis of the sales pipeline, and surface insights that enable smarter coaching of sales teams.

Dynamics 365 AI for Customer Service: With Microsoft’s AI and natural language understanding, customer service data can surface automated insights that help guide employees to take action and can even leverage virtual agents to help lower support costs and enable delightful customer experiences, all without needing in-house AI experts and without writing any code.

Dynamics 365 AI for Market Insights: Helps empower your marketing, social media and market research teams to make better decisions with market insights. Marketers can improve customer relationships with actionable web and social insights to engage in relevant conversations and respond faster to trends.

To help bring this to life, today we released a video with our CEO, Satya Nadella, and Navrina Singh, a member of our Dynamics 365 engineering team, showing examples of ways we’re bringing the power of AI to customer service organizations.

Dynamics 365 + Mixed Reality

Our second announcement of the day centers on the work we are doing to bring mixed reality and business applications together.

Since the release of Microsoft HoloLens over two years ago, the team has learned a lot from customers and partners. The momentum that HoloLens has received within the commercial space has been overwhelmingly positive. This has been supported by increased demand and deployment from some of the world’s most innovative companies.

We recognize that many employees need information in context to apply their knowledge and craft. Not only on a 2-D screen — but information and data in context, at the right place, and at the right time, so employees can produce even greater impact for their organizations. Mixed reality is a technology uniquely suited to do exactly that.

This is a whole new kind of business application. And that’s precisely what we’re introducing today, Dynamics 365 Remote Assist and Dynamics 365 Layout.

Today, we also showcased for the first time how Chevron is deploying HoloLens to take advantage of Dynamics 365 mixed reality business applications.

Chevron is already achieving real, measurable results with its global HoloLens deployment. Previously it was required to fly in an inspector from Houston to a facility in Singapore once a month to inspect equipment. Now it has in-time inspection using Dynamics 365 Remote Assist and can identify issues or provide approvals immediately.

In addition, remote collaboration and assistance have helped the company operate more safely in a better work environment, serving as a connection point between firstline workers and remote experts, as well as cutting down on travel and eliminating risks associated with employee travel.

Here is a peek into the work Chevron is doing with mixed reality:

Unlock what’s next with the Dynamics 365 October 2018 release

Next week at Microsoft Ignite and Microsoft Envision we’ll be in Orlando talking with thousands of customers, partners, developers, and IT and business leaders about our October 2018 release for Dynamics 365 and the Power platform that will be generally available Oct. 1. The wave of innovation this represents across the entire product family is significant, with hundreds of new capabilities and features.

We will have a lot more to talk about in the weeks and months ahead. We look forward to sharing more!

Tags: , , ,

AT&T Spark highlights big changes in networking market

SAN FRANCISCO — AT&T is revamping more than its massive network to deliver high-speed, low-latency 5G services to businesses and consumers. The company is also remaking its relationship with networking vendors.

At the heart of the change is an open-source-first policy that has redefined the role of tech vendors that historically supplied the service provider with proprietary hardware and software. Now, AT&T is telling its suppliers, which include Ericsson, Nokia and Samsung, they have to become software developers and system integrators.

“There is a place for all of those hardware vendors in this new ecosystem to become integrators, to become hardeners,” Amy Wheelus, vice president of AT&T’s Network Cloud, said this week during an interview at the service provider’s 5G AT&T Spark conference.

A hardener, which is a word Wheelus takes credit for, is a tech company that bolts features onto open source software, such as security and the management applications necessary to troubleshoot and fix network problems. Open source technology, such as OpenStack and Open Network Automation Platform (ONAP), is the foundation of AT&T’s Network Cloud, a cloud computing platform under construction to support future 5G applications.

Work like AT&T’s, which is also underway in other service providers’ data centers, is changing the relationship between carriers and networking suppliers, said Rajesh Ghai, an analyst at IDC.

“There’s a big architectural change that is happening in telcos today,” he said. “[And] there will be a huge systems integration and services component to it.”

The wish list from AT&T Spark

What AT&T and other service providers want matters to networking vendors. That’s because the suppliers’ largest customer base — enterprises — is gradually trading their on-premises software for applications running in the cloud. The trend is shrinking the enterprise market, while increasing the buying clout of cloud providers — such as Amazon, Google and Microsoft — and service providers preparing for massive 5G rollouts.

As a result, AT&T, Vodafone, Verizon and other large service providers “are going to be much more prescriptive in how we want [technology] to evolve,” Wheelus said.

For AT&T, that means commodity x86 hardware for running all Layer 4-7 network services, such as routing, load balancing and firewalls. AT&T is using ONAP to turn those services into virtualized network functions. Under the VNFs is the OpenStack cloud computing platform.

AT&T also wants makers of Layer 4-7 software to rethink the design of their products. For example, rather than selling load balancers for specific purposes, AT&T wants one product that it can configure for multiple tasks using open source orchestration tools, like Ansible and YAML.

“We don’t need eight different load balancers from eight different companies,” Wheelus said. “I need one load balancer, and then I fine-tune it.”

5G hype vs. reality

Is [there] a hype cycle? Yeah, maybe it’s a hype cycle.
Amy Wheelusvice president of AT&T’s Network Cloud

While revamping its network for 5G, AT&T is rolling out 5G radios and other infrastructure in U.S. cities. By early next year, AT&T plans to have 19 cities wired for 5G, including Atlanta, Dallas, Los Angeles and San Francisco.

What’s missing, however, are the applications that will deliver services that take advantage of 5G’s unique capabilities. At Spark, AT&T and its partners showed marketing videos touting the potential for better home gaming and entertainment and innovation in medicine and manufacturing. But none of the services exist, and it’s not clear how they will become a reality.

“Is [there] a hype cycle? Yeah, maybe it’s a hype cycle,” Wheelus said at AT&T Spark.

Nevertheless, AT&T claimed it is making progress on future products at its 5G development centers in Atlanta, Palo Alto, Calif., and Plano, Texas. The 5G infrastructure rolling out in the 19 cities is where AT&T will eventually test the services under development.

“We’re just getting standards-based equipment [for the cities],” Wheelus said.

With so much experimentation underway, no one can predict whether the billions of dollars AT&T and the rest of the tech industry are spending on 5G will generate a healthy return on investment. One of the first indicators will be the success of mobile services built for 5G-enabled smartphones, which analysts predict will arrive in 2021.

The potentially more lucrative services for businesses — the kind that is shown today in slick videos at 5G conferences — will take longer to hit the market. Those products are unlikely to be widely available until 2025, analysts said.

Box-IBM AI connection core component of Box partner strategy

SAN FRANCISCO — The Box-IBM partnership has spawned the first custom AI applications for the Box Skills Kit platform, codeveloped by IBM and the content management software vendor.

Box has been refining its Salesforce-style developer ecosystem by pursuing two tiers of partners: strategic partners with big market reach, like IBM, and application integrations with leading app vendors, like Salesforce, Facebook and Slack.

As for the strategic partnerships, which also include companies like Google and AT&T, “we work with like-minded companies that have the ability to bring us into their customer conversations,” Niall Wall, senior vice president for business development and partners at Box, said in interview at the BoxWorks 2018 user conference.

Box, IBM go to market together

In the Box-IBM partnership, the companies have formal go-to-market contracts in which IBM resells Box into its customer base, and the companies share about 400 customers.

Wall said a key differentiator for Box in these partnerships is that Box brings a distinct emphasis on cybersecurity and partners and customers can feel that their content is safe inside Box.

“We take all that complexity out,” Wall said. “We say to our customers, ‘We will guarantee that your data will remain secure as you are enhancing it with these third-party AI services.'”

In the latest development in the Box-IBM relationship, on Dec. 18, 2018, Box plans to commercially release three custom AI and machine learning-based Box Skills based on Watson technology and developed in conjunction with IBM.

Box Skills Kit and basic Box Skills also work with AI services from AWS, Google and Microsoft, though those vendors have not yet released custom applications.

Users field-tested AI skills

IBM Watson tested the Box content applications over the last year with Canadian bank ATB Financial and H&R Block Canada.

The AI skills are used to categorize, organize and search Box content using AI algorithms embedded in the Box Skills framework.

H&R Block is using the custom document insights skill, which uses natural language processing to parse text and process basic tax forms.

The tax preparation form uploads large tax documents. Then, Watson learns the language of the forms, identifies key metadata and information, and pulls it out.

“The tax specialist can now say, ‘Here’s all the information and what I need to do,’ versus a human being going in there and entering all this information,” said Rashida Hodge, vice president for embed and strategic relationships at IBM Watson.

Meanwhile, the Canadian bank is using the same document insights skill to analyze and extract information from home loan applications to make it easier for loan officers to handle instead of manually going through application forms.

The content layer in both of these Box-IBM use cases is the Box platform.

IBM talks Box

So, now, we have a foundation where data is living in Box, and we’re taking the complexity and obscurity of AI.
Rashida HodgeIBM Watson

“What I think is fantastic about Box is that, to start any AI project, what do you need? Data. So, now, we have a foundation where data is living in Box, and we’re taking the complexity and obscurity of AI,” Hodge said. “So, customers are saying, ‘We can get these kinds of insights and intelligence from our data in Box, and all we have to do is turn a switch on.'”

Another IBM-Box skill is the custom image insights skill, which uses optical character recognition to capture images and then tag and classify their metadata. A third skill automates document transcription translation into dozens of languages.

“Now, it’s more searchable and more consumable,” Hodge said. “When you talk about AI for the enterprise, the reason Box picked us for their launch partner for customization is it’s not about general tagging or general information about a document.”

“It’s about the specifics of a particular domain, and we can drive that metadata,” she said.

AI business outlook promising

Hodge said IBM was seeing robust sales lead activity at the conference, where David Kenny, IBM Watson senior vice president, joined Box CEO Aaron Levie on stage before thousands of Box users and where the IBM booth was getting heavy walk-up traffic.

Software engineers and salespeople at the booth on the busy partner exhibit floor handled a steady flow of inquisitive prospective buyers and other conference attendees.

One of these curious potential customers was Aali Hashmi, senior software architect at Box customer Fannie Mae, the government-backed mortgage lender.

Hashmi watched a demo and then engaged in a long, technical conversation with the IBM engineer, who then photographed with his smartphone Hashmi’s conference tag information to follow up after the conference.

In an interview, Hashmi said he can envision using the Box-IBM AI tools to digitize and auto classify content, like home mortgage forms and also HR documents.

“It looks promising,” he said.

Box security gets a boost with built-in Shield

SAN FRANCISCO — Box shops will have the ability to get granular with a new built-in Box security feature, but organizations will have to find a role for the tool alongside their other security platforms.

Box Shield, which was introduced at the file-sharing company’s annual conference, BoxWorks, will detect anomalies and risky user behavior within Box. Experts here discussed the potential behind Box Shield and how it might integrate with existing security and identity management tools within businesses.

“Security is such a tough problem,” said James Sinur, vice president at Aragon Research, based in Morgan Hill, Calif. “I haven’t found any security software that covers all aspects of it.”

How Box Shield works

Box Shield has three main functionalities: smart access, anomaly detection and a content firewall.

Where I think [Box] will make their contribution is by adjusting policies.
James Sinurvice president at Aragon Research

Smart access enables end users and IT admins to classify Box files according to their level of confidentiality. Then, IT admins can apply policies based on those classifications.

Anomaly detection helps IT to discover compromised accounts and identify access abuse. For example, if an end user accesses Box from Guatemala and downloads large amounts of data, Box Shield will flag that as risky behavior.

The content firewall feature can go beyond two-factor authentication to verify external users and check the security of devices.

IT can also use Box Shield to uncover historical data about a user’s activity and access analytics about their behavior.

Box Shield tries to play nice with other security

Sinur said he expects customers to use Box Shield in conjunction with other security platforms.

“Where I think [Box] will make their contribution is by adjusting policies that govern those pieces of [content],” he said.

Box is well-known for a plethora of integrations with third-party platforms — from Google and Slack to Microsoft and Okta. The company is already identifying places where Box Shield would integrate with other cloud access security broker (CASB) services, CEO Aaron Levie said in a press conference. Customers with an existing security information management tool, for example, would be able to use Box Shield in conjunction with it, he said.

An IT security analyst at a financial institution who wanted to remain anonymous was very interested in the new tool. His company already has several security technologies in place, such as Symantec and Okta, and would use Box Shield in addition to those services, he said.

“From a nonmanaged versus managed device, it would help us keep track of what’s going in and what’s going out based off of the device control,” he added.

Box Shield, however, would potentially replace the company’s current mobile device management platform, MobileIron.

“It would frequently push certificates out and start managing our CASBs,” he said. “We would use Box to help identify patterns in data movement.”

Pricing concerns

Pricing details aren’t yet released, but organizations will have to pay an additional cost for Box Shield, according to the vendor.

Pencils of Promise, a nonprofit organization in New York, is interested in Box Shield — but only at an affordable cost, said Ben Bromberg, senior manager of data systems at the nonprofit.

“It does seem like the sort of thing that an organization like mine would appreciate, but I have a suspicion that it would be at a price point that would be out of our reach,” he said.  

Box Shield will be available in private beta later this year, the company said.

Box Skills Kit to hit market soon in general release

SAN FRANCISCO — Box Inc. said it will launch a key piece of its Box Skills AI system, with expanded capabilities for customers to build and train their own AI-based content management tools in December 2018.

In addition to the expected commercial release of Box Skills Kit, the cloud content management vendor revealed new AI and workflow automation tools, Google integrations, security features and third-party app integrations at its BoxWorks 2018 conference.

A few hundred Box enterprise users have been using prefab Box Skills, the Box Skills Kit and Google integrations in private beta over the past year; the Google connections now are available in public beta.

Box also previewed a new Automations feature for workflow automation and a new data security system, Box Shield. Both are slated to be released in beta in early 2019, along with AI-driven Box Feed updates and notifications feature.

Box said Automations and an updated version of Box Tasks, with which users can assign things and deadlines to co-workers, will be out in beta in early 2019.

Activity Stream embeds third-party apps

All this came after Box unveiled Activity Stream, a collaboration system that enables users to work inside the Box platform with popular third-party apps like Slack, Salesforce and DocuSign. That product is expected to go into beta next year.

Users at the conference said they welcomed the open platform design of Activity Stream, Box’s progress on workflow automations, and the impending commercial availability of Box Skills Kit, which will include audio intelligence, video intelligence and image intelligence Skills, as well as the customization features.

The whole focus on workflow and integrations is really positive.
Rich LibbyCIO, Herbalife

“The whole focus on workflow and integrations is really positive,” said Rich Libby, CIO at natural supplements manufacturer Herbalife.

“I have all kinds of workflows that rely on Box documents where Box is not the final destination but stops along the track, so I’m excited to do that,” Libby said of using the Automations feature. “We have all kinds of contracts and agreements that we have to do offline or on email … so putting it on DocuSign will be great.”

Box highlights partner strategy

At the BoxWorks keynote, Box chief product and strategy officer Jeetu Patel  introduced speakers from Box partners like Apple, Google, Slack and ServiceNow.

BoxWorks executives meet media at BoxWorks 2018 conference
Box Inc. CEO Aaron Levie and Jeetu Patel, Box chief product officer, field media questions at BoxWorks 2018.

“In the digital workplace, what you need is a fundamentally modern set of tools to equip your users with so they can start innovating at a very different velocity,” Patel said of Box’s partner strategy. “For these tools to be effective they need to work seamlessly with your content. Regardless of the application that you the user may be working in, we want to make sure you can work with content in Box.”

Analysts said Box appears to be successfully executing a series of product advancements in a range of areas that users have been calling for, though some outstanding questions remain, particularly about the cost of Box Skills and Box Skills Kit.

The AI tools for developers work with AWS, Google, IBM and Microsoft AI, enabling users to customize and derive insights from Box content using those different engines. Users will apparently have to pay both the AI vendors for their services, as well as Box on a volume pricing basis.

Box Skills costs uncertain

“There is a lot of power in combining multiple Box features, like Skills and Automations,” said Alan Lepofsky, an analyst at Constellation Research who was at the conference at the George R. Moscone Convention Center.

“As the number of documents, images and videos stored in Box increases, the number of business use cases increases and the potential for AI goes up,” Lepofsky said. “What customers need to be aware of is these AI calls cost additional on top of Box.”

That cost does not appear to be fully worked out yet.

“The business model is actually quite simple for now. Skills Kit really just leverages our core platform, so it’s sort of a volume-based business model with the volume of data you’re moving back and forth between Box and third-party AI providers,” Box CEO Aaron Levie said in a Q&A session with reporters and analysts.

“We want to make it really, really easy to adopt Box Skills and make it easy to deploy it at scale, so we think it’s the fastest way customers are going to adopt this technology,” Levie said.

Beta user eyes commercial release

Meanwhile, a Box Skills beta user, Rich Guerra, head of application development at Farmers Insurance, said in an interview that he’s looking forward to using Box Skills Kit to develop AI applications for adjusters to enable them to work with videos and voice recording transcriptions from customers.

Farmers, which has standardized content on the Box platform, is also in the midst of a large-scale project to move legacy content from an old IBM on-premises file management system into the Box cloud.

As for Automations, “I was just texting our account rep how excited we were to take a look at that,” Guerra said. “The automation of the workflow is something we’re looking at to increase efficiency for adjusters, to make the process a lot easier.”

Box creating a cloud ecosystem

Overall, Box has done reasonably well in coordinating its far-flung product undertakings and positioning itself as a “Salesforce-like hub” of content services and collaboration, said Bola Rotibi, an analyst and research director at Creative Intellect Consulting.

“They’ve gotten a bit slicker than in the past,” Rotibi said of the orchestrated announcements on the first day of the conference.

“There are areas they can sharpen — like how much people are going to have to pay for Custom Skills — but I think they’re working behind the scenes to simplify that,” Rotibi said. “On the positive side, what I really like is the contextual integration of the third-party apps.”

Box Activity Stream embeds Salesforce, Slack in Box viewer

SAN FRANCISCO — With Box Activity Stream, the content management software vendor is bidding to make its cloud platform a collaboration hub for all Box users’ daily communications by integrating with popular third-party apps like Slack, Salesforce and DocuSign.

Unveiled at the BoxWorks 2018 conference here, Box Activity Stream enables users to use apps in the file preview pane of the Box user interface, where users tag each other about file sharing and exchange messages.

As well as giving users the ability to share and post links on non-Box apps, the new feature also recommends apps for people to use in conjunction with a file they are working on in Box. The app recommendations are customized according to how often a user chooses them, their popularity in the company, and the file type with which they are most frequently associated.

Announcement-beta cycle

Box Activity Stream is expected to see beta release next year, following a pattern of Box product releases being announced the year before they are available in beta.

Analysts familiar with Box Activity Stream said the technology is a useful addition to the Box platform, but that it also puts Box in the position of competing with a host of software platforms to be the go-to hub for enterprise users, and could also lead to notification overload.

“It helps Box go from cloud file storage to being an interactive user experience that involves content. It makes it a more collaborative workspace,” said Alan Lepofsky, an analyst at Constellation Research.

Vendors vying to be digital hub

“In theory, it’s a great concept,” Lepofsky added. “But everyone wants to be the digital hub. Everyone wants to fight for everybody’s attention and eyeballs and to do that they want to bring in all the other products.”

Everyone wants to be the digital hub.
Alan Lepofskyanalyst, Constellation Research

Meanwhile, the company views Box Activity Stream as a key part of its digital workplace strategy to redefine content management, workflow and services as digital first, said Faizan Buzdar, senior director of product management at Box.

While modern SaaS enterprise applications have accelerated time-to-market and time-to-adoption rates, they have also created a sort of scattering of content, Buzdar said.

“It’s an awesome trend, but at the same time it creates a challenge. How do I know what’s happening, how do I know where all that content lives?” he said.

Box Activity Stream enables users to, say, create a document in response to an email, send it to a collaborator for editing, send it to someone else over Slack and then attach it to an account in Salesforce or NetSuite.

With that process, Buzdar said, “our goal is to avoid content fragmentation and segmentation and let enterprises apply the same security and compliance layers across all their content from the perspective of the touch points that their end users have.”

Buzdar said Box has seen demand for this kind of capability among users in CRM, sales and ERP.

Screenshot of new Box Activity Stream feature
Box Activity Stream showing integrations with Slack, Salesforce and DocuSign

Google integrations

In addition to Box Activity Stream, Box on the first day of the conference said its previously announced Box for G Suite and Gmail integrations are now available for public beta use.

Box enterprise users have been calling for Google integrations more and more, Buzdar said.

“We love Google. We work closely with Google,” Buzdar said. “Customers are coming in who are basically deciding to standardize on Google. If you’re a big company, say with 100,000 employees, somewhere in the organization you have Google.”

The company also said Box Feed, which was announced at BoxWorks 2017, will also now go into public beta. The machine learning feature provides personalized updates, activities and recommended Box content.

A primitive precursor to Box Activity Stream was rolled out in 2011 when Box added a collaboration feature to its then mostly cloud document storage-focused platform, which it termed “activity streams.”

Mobile and desktop screenshots of Box Feed system
Box Feed displaying content trends, activity

Possible confusion, more engagement

As for Box Feed, Karen Hobert, a Gartner analyst, said with Box Activity Stream, Box runs the risk of confusion between the two.

“One would think a user might want them combined as long as they could control the experience. But maybe the different UI experiences — Activity Stream in viewers, Feed in Box UI — will mitigate any confusion,” Hobert said.

But Hobert said she sees value in Box Activity Stream in terms of smoothing what can be a sometimes disjointed experience toggling between apps and Box.

“Basically, I see it as a way to keep employees engaged in Box throughout the day. Certainly users will like not having to bounce around from app to app,” she said. “Activity Stream clearly makes a more seamless experience with Box and content in other apps. In the end, Box wants to — and needs to — be a destination that users won’t live without.”

Hobert also questioned Box’s record on delivering on new systems and features, noting that there were “significant delays” in two earlier products, Box Relay and Box Sync, and that Box Feed and Box Skills, the company’s high-profile AI play that was announced nearly a year ago, are still in beta.

Lepofsky said he expected significant news at the conference about the much-touted Box Skills system.

“Otherwise, they’re going to look bad,” he said.

Where the IBM Project Debater AI system may be headed

Last month, in almost real time and in front of a live audience in San Francisco, two humans debated facts and ethics with an AI system, IBM Project Debater.

The AI platform offered points and rebuttals during its first public debate, going first against Noa Ovadia, the 2016 Israeli national debate champion, on the issue of subsidizing space exploration, and then against Israeli professional debater Dan Zafrir on the topic of telemedicine. Debaters, human and machine, were not made aware of the subjects ahead of time.

According to a snap poll of the audience after each session, audiences felt that Project Debater, at least on the topic of space exploration, enriched their knowledge more than its human counterpart.

The humans, however, were largely found to be better, more persuasive speakers. Indeed, despite its digital prowess, some debate experts have noted that the AI debating system lacks a certain ability to deploy tonal effects, such as irony and sarcasm.

The debating system isn’t IBM’s first foray into machine-human jousting. In 2011, IBM’s Watson supercomputer beat trivia stars in a game of Jeopardy, and in 1997, IBM’s Deep Blue chess computer bested world chess champion Garry Kasparov.

[embedded content]

IBM Research teaser for Project Debater

The Project Debater engineers appear to be taking a more nuanced approach this time.

Ranit Aharonov, manager of the IBM Project Debater team, said members of the project team didn’t think of the system simply as something that could win a debate.

“When we look at a debate, we don’t only look at who swayed the audience more. There’s a lot more to it,” Aharonov said.

Building a debating machine

Developed over the past six years at the IBM Research lab in Haifa, Israel, IBM Project Debater uses sophisticated machine learning algorithms and millions of newspapers and articles to identify and organize facts relevant to a debate topic.

The AI is able to cluster that information into themes based on the topic of debate, and, using what IBM calls data-driven speech writing, delivers the information in a coherent sentence.

IBM Project Debater is essentially trained in the art of debate — to have a general idea of when and how to use factual and ethical arguments to support or dispute a point. To be able to debate in real time, the system employs natural language processing to identify the main components of an opponent’s speech and then give a rebuttal.

While pursuing this researching project, during the process, we are actually finding ourselves facing new problems we haven’t faced before.
Noam SlonimIBM

According to Noam Slonim, principal investigator for IBM Project Debater, the technology could have a number of applications. The most immediate one, he said, is advancing the field of science.

“While pursuing this researching project, during the process, we are actually finding ourselves facing new problems we haven’t faced before,” Slonim said

Real-life applications

Beyond advancing the field, Slonim said he sees “the underlying technology with the Debater being very, very aligned with technologies that help people make a better-informed decision,” noting that it could eventually have uses in the fields of politics or business.

“Just imagine giving Debater a topic and asking it to find everything of relevance to that topic and what that could mean,” he said.

Also, Slonim said he sees the technology being useful in the education field. Having a debating AI system “can help kids learn how to build better arguments and become more informed in a topic.”

Dan Zafrir and IBM Project Debater, San Francisco
Israeli debater Dan Zafrir poses with IBM’s Project Debater before a public debate in San Francisco last month

“IBM Project Debater, while still in development, could be brought into specific use cases, and is slated to be released in some form next year,” Slonim said. He declined to say what incarnation the technology might take.

As for bringing some of the technology behind IBM Project Debater to IBM’s well-publicized AI system, Watson, Slonim said: “The implication is that these will be incorporated into Watson and enhance its capabilities.”

Mixed review from an analyst

Adrian Bowles, vice president of research and lead analyst for artificial intelligence at Aragon Research, was at last month’s live debate in San Francisco.

Bowles, who said he first spoke with an IBM representative about the IBM Project Debater four years ago, said he was struck more by how IBM Project Debater identified arguments than how it expressed them.

“The natural language generational software is not nearly as impressive to me as what they’ve done with natural language understanding,” he said, adding that the AI system presented arguments more on a high school or college level than a professional one.

“Finding and representing the logical position and being able to identify the opposite of that is where the magic happens, if you will,” Bowles continued.

Bowles agreed that the technology could be useful when applied in a classroom setting, but noted that he would also like to see it used to help extract provable facts from bodies of text, like multiple sources of news.

Specifically, Bowles cited fake news and the political bias reflected by news sources. Technology in IBM Project Debater could be used to analyze multiple news sources on the same topic and help separate facts from bias or misreporting.

“What I would like to see is it being able to identify practical arguments and being able to map those out,” he said.

Due to the vast number of documents Debater has access to, Slonim said he would expect technology to be separated out before the system is commercialized, partitioning it out to get to the basic underlying technology and allowing users to input their own data to be analyzed.

“I think that would get people using it and experimenting in novel ways,” he said.

Databricks platform additions unify machine learning frameworks

SAN FRANCISCO — Open source machine learning frameworks have multiplied in recent years, as enterprises pursue operational gains through AI. Along the way, the situation has formed a jumble of competing tools, creating a nightmare for development teams tasked with supporting them all.

Databricks, which offers managed versions of the Spark compute platform in the cloud, is making a play for enterprises that are struggling to keep pace with this environment. At Spark + AI Summit 2018, which was hosted by Databricks here this week, the company announced updates to its platform and to Spark that it said will help bring the diverse array of machine learning frameworks under one roof.

Unifying machine learning frameworks

MLflow is a new open source framework on the Databricks platform that integrates with Spark, SciKit-Learn, TensorFlow and other open source machine learning tools. It allows data scientists to package machine learning code into reproducible modules, conduct and compare parallel experiments, and deploy models that are production-ready.

Databricks also introduced a new product on its platform, called Runtime for ML. This is a preconfigured Spark cluster that comes loaded with distributed machine learning frameworks commonly used for deep learning, including Keras, Horovod and TensorFlow, eliminating the integration work data scientists typically have to do when adopting a new tool.

Databricks’ other announcement, a tool called Delta, is aimed at improving data quality for machine learning modeling. Delta sits on top of data lakes, which typically contain large amounts of unstructured data. Data scientists can specify a schema they want their training data to match, and Delta will pull in all the data in the data lake that fits the specified schema, leaving out data that doesn’t fit.

MLflow's tracking user interface
MLflow includes a tracking interface for logging the results of machine learning jobs.

Users want everything under one roof

Each of the new tools is either in a public preview or alpha test stage, so few users have had a chance to get their hands on them. But attendees at the conference were broadly happy about the approach of stitching together disparate frameworks more tightly.

Saman Michael Far, senior vice president of technology at the Financial Industry Regulatory Authority (FINRA) in Washington, D.C., said in a keynote presentation that he brought in the Databricks platform largely because it already supports several query languages, including R, Python and SQL. Integrating these tools more closely with machine learning frameworks will help FINRA use more machine learning in its goal of spotting potentially illegal financial trades.

You have to take a unified approach. Pick technologies that help you unify your data and operations.
John Golesenior director of business analysis and product management at Capital One

“It’s removed a lot of the obstacles that seemed inherent to doing machine learning in a business environment,” Far said.

John Gole, senior director of business analysis and product management at Capital One, based in McLean, Va., said the financial services company has implemented Spark throughout its operational departments, including marketing, accounts management and business reporting. The platform is being used for tasks that range from extract, transform and load jobs to SQL querying for ad hoc analysis and machine learning. It’s this unified nature of Spark that made it attractive, Gole said.

Going forward, he said he expects this kind of unified platform to become even more valuable as enterprises bring more machine learning to the center of their operations.

“You have to take a unified approach,” Gole said. “Pick technologies that help you unify your data and operations.”

Bringing together a range of tools

Engineers at ride-sharing platform Uber have already built integrations similar to what Databricks unveiled at the conference. In a presentation, Atul Gupte, a product manager at Uber, based in San Francisco, described a data science workbench his team created that brings together a range of tools — including Jupyter, R and Python — into a web-based environment that’s powered by Spark on the back end. The platform is used for all the company’s machine learning jobs, like training models to cluster rider pickups in Uber Pool or forecast rider demand so the app can encourage more drivers to get out on the roads.

Gupte said, as the company grew from a startup to a large enterprise, the old way of doing things, where everyone worked in their own silo using their own tool of choice, didn’t scale, which is why it was important to take this more standardized approach to data analysis and machine learning.

“The power is that everyone is now working together,” Gupte said. “You don’t have to keep switching tools. It’s a pretty foundational change in the way teams are working.”

VMware is redesigning NSX networking for the cloud

SAN FRANCISCO — VMware is working on a version of NSX for public clouds that departs from the way the technology manages software-based networks in private data centers.

In an interview this week with a small group of reporters, Andrew Lambeth, an engineering fellow in VMware’s network and security business unit, said the computing architectures in public clouds require a new form of NSX networking.

“In general, it’s much more important in those environments to be much more in tune with what’s happening with the application,” he said. “It’s not interesting to try to configure [software] at as low a level as we had done in the data center.”

Four or five layers up the software stack, cloud provider frameworks typically have hooks to the way applications communicate with each other, Lambeth told reporters at VMware’s RADIO research and development conference. “That’s sort of the level where you’d look to integrate in the future.”

Todd Pugh, IT director at Sugar Creek Packing Co., based in Washington Court House, Ohio, said it’s possible for NSX to use Layer 7 — the application layer — to manage communications between cloud applications.

“If we burst something to the cloud on something besides AWS, the applications are going to have to know how to talk to one another, as opposed to just being extensions of the network,” Pugh said.

Today, VMware is focusing its cloud strategy on the company’s partnership with cloud provider AWS. The access VMware has to Amazon’s infrastructure makes it possible for NSX to operate the same on the cloud platform as it does in a private data center. Companies use NSX to deliver network services and security to applications running on VMware’s virtualization software.

Pugh would not expect an application-centric version of NSX to be as user-friendly as NSX on AWS. Therefore, he would prefer to have VMware strike a similar partnership with Microsoft Azure, which would give him the option of using the current version of NSX on either of the two largest cloud providers.

“I can shop at that point and still make it appear as if it’s my network and not have to change my applications to accommodate moving them to a different cloud,” Pugh said.

Nevertheless, having a version of NSX for any cloud provider would be useful to many companies, said Shamus McGillicuddy, an analyst at Enterprise Management Associates, based in Boulder, Colo.

“If VMware can open up the platform a bit to allow their customers to have a uniform network management model across any IaaS environment, that will simplify engineering and operations tremendously for companies that are embracing multi-cloud and hybrid cloud,” McGillicuddy said.

VMware customers can expect the vendor to roll out the new version of NSX over the next year or so, Lambeth said. He declined to give further details.

Rethinking NSX networking

More lately, I’ve been sort of taking a step back and figuring out what’s next. I feel like the platform for NSX is kind of in a similar situation to where ESX and vSphere were in 2006 and 2007.
Andrew Lambethengineering fellow at VMware

VMware will have to prepare NSX networking, not just for multiple cloud environments, but also the internet of things, which introduces other challenges to network management and security.

“More lately, I’ve been sort of taking a step back and figuring out what’s next,” Lambeth said. “I feel like the platform for NSX is kind of in a similar situation to where ESX and vSphere were in 2006 and 2007. Major pieces were kind of there, but there was a lot of buildout left.”

VSphere is the brand name for VMware’s suite of server virtualization products. ESX was the former name of VMware’s hypervisor.

VMware’s competitors in software-based networking that extends beyond the private data center include Cisco and Juniper Networks. In May, Juniper introduced its Contrail Enterprise Multicloud, while Cisco has been steadily developing new capabilities for its architecture, called Application Centric Infrastructure.

The immediate focus of the three vendors is on the growing number of companies moving workloads to public clouds. Synergy Research Group estimated cloud-based infrastructure providers saw their revenue rise by an average of 51% in the first quarter to $15 billion. The full-year growth rate was 44% in 2017 and 50% in 2016.

Juniper Contrail battles Cisco ACI, VMware NSX in the cloud

SAN FRANCISCO — Juniper Networks has extended its Contrail network virtualization platform to multicloud environments, competing with Cisco and VMware for the growing number of enterprises running applications across public and private clouds.

The Juniper Contrail Enterprise Multicloud, introduced this week at the company’s NXTWORK conference, is a single software console for orchestrating, managing and monitoring network services across applications running on cloud-computing environments. The new product, which won’t be available until early next year, would compete with the cloud versions of Cisco’s ACI and VMware’s NSX.

Also at the show, Juniper announced that it would contribute the codebase for OpenContrail, the open source version of the software-defined networking (SDN) overlay, to The Linux Foundation. The company said the foundation’s networking projects would help drive OpenContrail deeper into cloud ecosystems.

Contrail Enterprise Multicloud stems, in part, from the work Juniper has done over several years with telcos building private clouds, Juniper CEO Rami Rahim told analysts and reporters at the conference.

“It’s almost like a bad secret — how embedded we have been now with practically all — many — telcos around the world in helping them develop the telco cloud,” Rahim said. “We’ve learnt the hard way in some cases how this [cloud networking] needs to be done.”

Is Juniper’s technology enough to win?

Technologically, Juniper Contrail can compete with ACI and NSX, IDC analyst Brad Casemore said. “Juniper clearly has put considerable thought into the multicloud capabilities that Contrail needs to support, and, as you’d expect from Juniper, the features and functionality are strong.”

Cisco and VMware have marketed their multicloud offerings aggressively. As such, Juniper will have to raise and sustain the marketing profile of Contrail Enterprise Multicloud.
Brad Casemoreanalyst, IDC

However, Juniper will need more than good technology when competing for customers. A lot more enterprises use Cisco and VMware products in data centers than Juniper gear. Also, Cisco has partnered with Google to build strong technological ties with the Google Cloud Platform, and VMware has a similar deal with Amazon.

“Cisco and VMware have marketed their multicloud offerings aggressively,” Casemore said. “As such, Juniper will have to raise and sustain the marketing profile of Contrail Enterprise Multicloud.”

Networking with Juniper Contrail Enterprise Multicloud

Contrail Enterprise Multicloud comprises networking, security and network management. Companies can buy the three pieces separately, but the new product lets engineers manage the trio through the software console that sits on top of the centralized Contrail controller.

For networking in a private cloud, the console relies on a virtual network overlay built on top of abstracted hardware switches, which can be from Juniper or a third party. The system also includes a virtual router that provides links to the physical underlay and Layer 4-7 network services, such as load balancers and firewalls. Through the console, engineers can create and distribute policies that tailor the network services and underlying switches to the needs of applications.

Contrail Enterprise Multicloud capabilities within public clouds, including Amazon Web Services, Google Cloud Platform and Microsoft Azure, are different because the provider controls the infrastructure. Network operators use the console to program and control overlay services for workloads through the APIs made available by cloud providers. The Juniper software also uses native cloud APIs to collect analytics information. 

Other Juniper Contrail Enterprise Multicloud capabilities

Network managers can use the console to configure and control the gateway leading to the public cloud and to define and distribute policies for cloud-based virtual firewalls.

Also accessible through the console is Juniper’s AppFormix management software for cloud environments. AppFormix provides policy monitoring and application and software-based infrastructure analytics. Engineers can configure the product to handle routine networking tasks.

The cloud-related work of Juniper, Cisco and VMware is a recognition that the boundaries of the enterprise data center are being redrawn. “Data center networking vendors are having to redefine their value propositions in a multicloud world,” Casemore said.

Indeed, an increasing number of companies are reducing the amount of hardware and software running in private data centers by moving workloads to public clouds. Revenue from cloud services rose almost 29% year over year in the first half of 2017 to more than $63 billion, according to IDC.