How AI will change enterprise mobility

How AI will change enterprise mobility

Apps, device management and the user experience will all be affected. But the biggest developments could be in the security area.

By Bob Violino for Computerworld

Your smartphone is about to get smarter, thanks to artificial intelligence (AI) and machine learning (ML). And that has huge implications for enterprise support for mobility.

Enterprise mobility has long promised to allow workers to be productive wherever they are, to speed up business processes and to improve accuracy and efficiency by putting the most up-to-date data in the hands of workers in the field, says Kevin Burden, vice president of mobility research and data strategy at 451 Research. The addition of AI will help deliver on those promises, he says.

The ways it will do that are multifaceted, with the effects seen in the areas of device management, user experience, security, applications and the very devices themselves. At the same time, new concerns about privacy are sure to arise as AI and ML become ever more efficient at gathering data points.

“AI is going to mean new applications and even possibly new device types, primarily because AI will alter and improve the business logic within apps,” Burden says. Applications will be able to take advantage of advanced user interfaces with speech and visual gesture recognition.

“One element of enterprise mobility that will clearly benefit from AI is the organizational challenges that were created by having a disparate and mobile workforce,” Burden says. Application providers will apply ML to user activity streams, giving organizations insight into how end users spend their time, he says. As patterns of behavior are identified, organizations will be able to improve processes and the user experience.

Easier authentication is one example. Pattern recognition is an AI strength. Because AI can gather huge amounts of such data and recognize anomalies with ease, it can make authentication much more transparent for users, says Chris Silva, research vice president for enterprise mobile strategy at Gartner.

Some of the more advanced algorithms detect how a user enters text and analyze their gait. Pair those distinctive patterns with information on the user’s active connections and GPS data, Silva says, and “the number of layers of multifactor authentication or constant requirements to enter passwords could be greatly reduced.”

Another AI/ML improvement will be in speech-to-text capabilities, allowing that technology to replace smartphone data input in some situations, says Phil Hochmuth, program director for enterprise mobility at research firm IDC. “Verticals such as medical and others will use speech for data input for basic tasks such as records and workflow updates,” Hochmuth says.

And, he says, applications will become intuitive in whole new ways: “ML will also be integrated more into mobile applications to enable quicker decisions, responses and inputs to anticipate user actions, as opposed to requiring users to look for options in windows and dropdowns.”

IT will benefit from AI’s and ML’s assistance with device management. For example, Silva says, the technology can be used to scan all of the devices in an organization and proactively notify the administrator of issues, such as the discovery that 25% of the organization’s Android devices are two versions out of date. Even more helpful for IT organizations that are short of personnel is the potential to automate actions based on the information discovered by AI/ML, Silva says. The technology will really pay off for IT once the systems can use AI to detect and remediate issues on the fly, he says.

IT is also likely to appreciate many of the AI-fueled user-experience enhancements that are coming to email, contact and calendar tools as vendors add personal-assistant technology. It’s fairly common already for calendars to use AI to tell users when they should leave for an appointment.

The advantage to IT isn’t direct, but many IT departments want users to stick to their company-provided email, contact and calendar tools when working, Silva notes, as a way to protect and segregate work data from personal. The new user-facing convenience features could make using those tools more appealing to users.

While it’s still unclear how AI will impact the overall mobility market on a long-term basis, it is certain that “the EMM [enterprise mobility management] space is very crowded, without any real significant differentiation,” says 451 Research’s Burden, so vendors will look to AI for new ways to innovate.

AI and security

Perhaps the area with the greatest potential to get a boost from AI, and particularly its pattern-recognition chops, is security. Certainly many vendors are already incorporating AI/ML in their security offerings as a way to boost performance.

One area where vendors already have offerings is ML-based mobile threat detection. For example, MobileIron uses ML in its new MobileIron Threat Defense tool, which employs usage and behavioral analysis to detect suspicious behaviors in mobile apps or networks and then learns from the information it gathers to continuously improve its ability to detect malware and rogue networks.

Sophos has integrated deep learning into its endpoint security products that provide what it calls “predictive security.” The company aims to extend this deep learning layer to all endpoints, including mobile ones. It has also introduced an email protection tool that uses the same technology to intercept more threats before they can make it onto the endpoints.

Other vendors see an opportunity to use AI to help IT departments that are stretched thin to make sense of all the data that is gathered by their existing endpoint management tools. Among them is Citrix, whose unified endpoint management offering also manages all devices that enter the workplace, including laptops, mobile phones, tablets and wearables. The Citrix security analytics application monitors those devices and helps IT to apply security policies and ensure that the network remains secure.

Citrix Analytics also performs user-behavior analytics, applying machine learning to categorize users as high, medium or low risks and then adjusting the risk scores as more data comes into the system.

IBM, meanwhile, has developed MaaS360 with Watson, a cloud-based application designed to help IT administrators make sense of the massive amounts of data generated by endpoints and their users, apps and content. It applies cognitive technologies to security, end-user productivity, mobile app management and administration.

EMM users are inundated with more information than they can absorb about apps, configuration/policy best practices, productivity tools, and emerging threats and vulnerabilities, IBM explains. IBM MaaS360 delivers cognitive insights, embedded in the platform, to help organizations wade through the information they’re gathering and distill it into insights and recommendations that are relevant to their business. The core of MaaS360 is IBM Watson technology, which can index and annotate huge volumes of datasets to look for relevant data that applies contextually to each individual client deployment of MaaS360.

A privacy backlash?

One dark cloud on the AI/ML front is data privacy.

Users have become more aware of the perils of their personal information ending up in the hands of companies such as Facebook and Google, says Gartner’s Silva. So “the idea of an employer or other company retaining the outputs of their mobile devices, apps and data usage — which Gartner calls workplace analytics — is sure to meet opposition from some users,” he says.

Those concerns can’t be ignored, especially given the emergence of strict regulations such as the European Union’s General Data Protection Regulation (GDPR) and the California Consumer Privacy Act of 2018.

Those regulatory concerns could strip the utility from mobile offerings dependent on AI/ML, Silva says. While user pushback on data will not negate the value of AI/ML in mobile offerings, it could impede the collection of data for some or all users, Silva says. “That, in turn, could make the data less useful for some groups of users or some regions, while still providing value to others,” he says.

To ameliorate this, organizations should be forthright in discussing what data they collect and how it will be used, Silva says. Gartner advises clients to illustrate the outcome and its benefit to users and take pains to note what won’t be collected or done with data. “The list of what IT does not do with data should almost always be longer than the list of what it does or can do with data,” he says.

Read the original article online at Computerworld.