photo illustration of AI
Artificial intelligence holds plenty of promise — but it's "not magic," one expert says. PHOTO: petcor80

The term artificial intelligence (AI) was coined in 1955 when a small group of scientists lobbied the Rockefeller Foundation to fund the Dartmouth Workshop, a two month project whose purpose would be "to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves."

It is the same definition that many of us use today. So what exactly have we been doing for the last 60 years?

Despite its age, these are still early days for AI, according to Andrew Ng, founder of the Google Brain project at Google, the former director of the Stanford Artificial Intelligence Laboratory and chief scientist at Baidu Research. 

AI, he states unequivocally in an article in Harvard Business Review, "will transform many industries." 

"But it’s not magic," he adds. "If a typical person can do a mental task with less than one second of thought, we can probably automate it using AI either now or in the near future," he wrote.

OK, so it's no silver bullet. But it's already moved beyond a fad to the foundation of useful products. 

In a recent CMSWire story, Amit Kothari, CEO of Tallyfy, said the biggest developments of AI in the office can be split into two distinct areas.
  • The first is natural language processing (NLP), which is the capture and identification of data
  • The second is machine learning, the ability to see data patterns and predict outcomes through analyzing historical patterns

He predicts interesting things: "We need to think of AI as a tool for coaching, in this way we can tap into human expertise — and enhance this with machines."

The Question

"What's the next step in the evolution of AI?"

The Answers

John Ball, SVP, Salesforce and GM, Salesforce Einstein

John Ball
John Ball
Ball oversees the innovation and product direction of delivering AI to Salesforce's Customer Success Platform. Before his current role, he was the CEO of KXEN for four years, where he lead its transition from a tools company to a predictive applications company. Before joining KXEN, Ball served in several executive roles at Salesforce. During a five-month sabbatical in 2014, he was Chief Fun Officer, "shuttling kids, cooking, swimming and biking, marathon TV series binges." Tweet to John Ball

In 2016, the industry made incredible progress to democratize the power of artificial intelligence even further. AI interfaces and technology are now so intuitive and easy to use that they have become a seamless part of the way we run our personal lives.

But there is still a lot of work to do when it comes to bringing AI into our businesses. The challenge is twofold: getting AI right is hard and the vast majority of companies just don't have the technical expertise and infrastructure required. In addition, AI should be seamlessly integrated into the business user experience. Until recently, the insights, predictions and recommendations coming from AI were available in a separate application and only accessible by data experts, not the business users.

The next evolution of AI will ring in the age of the citizen developer where every business user — not PhDs and data scientists — are able to leverage AI to make smarter, more predictive decisions. 

Intelligent business apps will become more consumer in nature, mimicking the simple experience and lowering the barriers to entry to empower business users across sales, service, marketing and IT to transform their business. Just like Siri is built into our phones and Netflix recommendations appear right on the homepage, AI will be embedded directly where people work — not siloed in complicated tools that are controlled by a select few. 

Predictive insights and recommended best next actions will appear within daily business workflows and every customer interaction will become more intelligent, and in turn, every business user will become more productive. 

Steve Lucas, CEO, Marketo

Steve Lucas
Steve Lucas
Lucas won the top job at Marketo just three months ago, leaving a career at SAP behind him. At SAP he was most recently the president of platform solutions, leading the organization's growth and expansion into a range of technology markets. Before that he was global executive vice president and GM for its database and middleware business, as well as global GM for its business analytics organization. Before SAP, Lucas was SVP of platform marketing at Salesforce. Tweet to Steve Lucas.

We reap all the benefits from analytics, computational insights, predictive models, and the like, especially when predictability provides us with assistance or convenience. Cut to the chase: humans are willing to behave predictably for the sake of convenience.

To that end, I believe we will see a dramatic acceleration of adaptive and intuitive applications and technologies, such as AI, unlike we have ever seen. 

This acceleration will make all of our devices and applications smarter in order to accommodate our behaviors. And in a world where technology is ubiquitous and everyone and everything is connected — a notion I call the "Engagement Economy" — the winners will use these technologies and more to engage with people in an authentic and personalized way at scale.

Jamie Anderson, Senior Vice President and CMO, SAP Hybris

Jamie Anderson
Jamie Anderson
Anderson is responsible for leading solution and industry marketing across the SAP Hybris solutions portfolio. Before this, he led global solutions marketing for SAP's Customer Line of Business (LoB) solutions (CRM) and was also responsible for positioning and messaging development for SAP's LOB Customer (Sales, Service, and Marketing) solutions. Anderson also spent time at Adobe as head of its Financial Services Industry solutions marketing in Europe, the Middle East and Africa and at Siebel as a principal solutions consultant, financial services. Anderson was formerly a singer/songwriter with Scottish rock band Big Wednesday. Tweet to Jamie Anderson.

Today's conversations are simplifying the potential of AI by putting it in a narrow scope. The market discusses AI generally as an automated program response (chatbots), but it can be enabled to learn and adjust responses based on the questions it is asked. 

The future of AI lies in pairing the technology with real-time, live data to create personalized and contextualized experiences. AI has the ability to go beyond machine learning by interpreting and understanding the subtext of human emotion to create engaging experiences.

In this, AI can completely change the customer experience because most consumers make decisions based on its emotional impact. By interpreting explicit signals that are usually only visible during in-person conversations, such as facial impressions, marketers can truly understand their customers and provide intelligent responses.

With facial recognition we are close to harnessing this power, but the power of AI lies in using every single piece of data you have access to. HBO's recent show, Westworld, brought up the emotional capability of machines. 

Compared to the 1970's version of Westworld, it's easier to imagine a world where we interact with AI on a daily basis because of the data processing power we now have access to. Having access to mass amounts of real-time data is what makes AI functional and effective today, but we need to continue to add data to make it smarter and more impactful.

Noah Wasmer, GM and SVP, Mobile Products, VMware

Noah Wasmer
Noah Wasmer
Wasmer is responsible for global research and development for VMware's mobile product lines including AirWatch EMM and next-gen mobile productivity apps. Before rejoining VMware in 2014, Wasmer was iPad product marketing manager at Apple. He says he is "motivated to create delightful products that help people love their work." Tweet to Noah Wasmer.

VMware has already integrated AI technology into some of its end-user computing solutions such as VMware AirWatch Unified Endpoint Management, where the solution automates daily mundane administrative tasks to reduce the size and burden on IT required to manage hundreds of thousands of devices and allow them to focus on other priorities.

The new world of unified endpoint management (UEM) backed by intelligent analytics enables next gen Win 10 PCs, mobile devices, wearables and even IoT to be updated, patched and secured in real time — while providing a user experience that delivers a real-time catalog of services.

What makes it possible? Analytics and intelligence. In next gen mobile and cloud operating systems (iOS, Android, Chromebooks, Win 10, etc.) a fundamental root of trust is established (and secured in a secure enclave) provides an analytics engine the ability to receive near real-time posture and context data — imagine 700 points of data every 30 milliseconds. 

By analyzing the data and leveraging prediction models such as tensor flow, true business problems can be handled easily and automated. Security is one area where we can make improvements in automation using AI with user/peer group learning while still maintaining user privacy.

For example, if an employee jail breaks or "roots" a mobile device, rather than a full erase of the device, a UEM solution with integrated artificial intelligence can instantaneously remove all corporate data and apps while leaving the personal data intact.

As businesses assess global threat intelligence, real-time information provides quick data points to elevate security. If an endpoint device is detected as compromised in a specific office location, artificial intelligence can make the decision to automatically prompt other users who visited that area in the past 30 days for additional authentication while the incident is being reviewed and mitigated.