standing on the edge

It has been clear for some time that artificial intelligence has already evolved and is entering a new phase of development. While it is not entirely clear as yet how, or even if, enterprises will use this new advanced AI, before looking at new strategies for this cutting edge technology — and edge is a key word here — it is important to understand what is and what it can possibly do.

In its recent Hype Cycle for emerging technologies (subscription required) and subsequent analysis of where the evolution of digital innovation is going, Stamford, Conn.-based, Gartner, took a look at advanced analytics and its possible role in the enterprise. According to the research, advanced analytics is the autonomous or semiautonomous examination of data or content using sophisticated techniques and tools, typically beyond those of traditional business intelligence (BI).

In a statement about the research, Brian Burke, research vice president at Gartner explained: “The adoption of edge AI is increasing for applications that are latency-sensitive (e.g., autonomous navigation), subject to network interruptions (e.g., remote monitoring, natural language processing [NLP], facial recognition) and/or are data-intensive (e.g., video analytics).”

The technologies to track include adaptive machine learning (ML), edge AI, edge analytics, explainable AI, AI platform as a service (PaaS), transfer learning, generative adversarial networks and graph analytics.

Related Article: 7 Ways Artificial Intelligence is Reinventing Human Resources

Adopting Advanced Analytics

Over the last decade, there has been a significant increase in adoption of advanced analytics across the enterprise, Shawn Rogers, senior director of analytic strategy at Palo Alto, Calif.-based Tibco, told us. By using advanced analytics with newer iterations of AI technology, companies are able to achieve a level of maturity in how they use data to generate innovation.

“It’s important to note,” he said “that AI has actually been around for a while, but it is now being leveraged in new ways that add value to advanced analytics technologies. “

Combining advanced analytics with AI, companies now look to bring analytics closer to their data rather than the more traditional approach of taking the data back to analytics. This agile approach lends itself to AI deployed at the edge and helps technology meet the most pressing demands of any enterprise organization — the ability for data-driven decisions to be made at the speed of business and in near-real-time. “These recent advances in technology, when combined with a strong foundation, allow companies to address the many data-intensive use cases found in today’s enterprise,” he added.  “IoT is an excellent example of predictive analytics providing great value at the edge, especially when its capabilities are combined with AI models to give companies business insights that weren’t previously accessible.”

Related Article: Why Artificial Intelligence Will Create More Jobs Than it Destroys

Advanced Analytics And Privacy

There is another use for advanced analytics in the enterprise that is becoming increasingly important and which is likely to grow and spur adoption in the future, notably data privacy. As enterprises in the EU and NA work to comply with data privacy regulations, they’re finding themselves at a disadvantage against regions that don’t have such policies in place. The challenge for enterprises that must conform to data privacy regulations becomes maintaining an internet that is intelligent, compliant and facilitates the unimpeded sharing of information at scale, Andrew Flip Flipowski, co-founder of Fluree, a Winston-Salem, North Carolina-based data management platform that provides a blockchain graph database.

As AI and advanced analytics continue to extend to the edge, these enterprises must question whether they are establishing network governance and an infrastructure that can accommodate the scalable free flow of data required to power these tools at scale and across geopolitical and geographical borders.

The semantic web — an aspect of Web 3.0 as defined by Tim Berners Lee as far back as 2006 – emphasizes the development of computing infrastructure and general internet governance capable of enabling these kinds of technologies to operate seamlessly across networks, and, more importantly, across geographic borders, facilitating machine to machine communication and interplay. *

The technology needed to optimize analytics while adhering to regulations is advanced enough to be considered Web 3.0, and isn't easily attainable. Some pieces of the Web 3.0 puzzle are just now entering the market to provide the necessary data platform and their unavailability has hampered the near-term deployments that have been envisioned for well over a decade. Now finally, we have graph, blockchain and machine-defined query capability to get to Web 3.0 nirvana.

For Web 3.0 to become standard, all enterprises must comply with data privacy regulations. Until then, enterprises without regulations will remain at an inherent advantage — with the caveat that if they want to do business globally, they, too, will face compliance hurdles.

Since competency and success is inversely proportional to privacy, the war for AI supremacy has already been abdicated by those with the greatest privacy regulations. Bottom line is: China won and EU and NA has lost. Only those with unencumbered access to all data regardless of sensitivity can expect to develop killer AI technology.

AI Moving To The Center

AI is taking on a more central role in the enterprise—in the same way AI is foundational to areas such as self-driving cars, automated financial trading and so on, Wilson Raj, global director of customer intelligence for SAS, added. He cites the example of the marketing function in the enterprise.

AI provides tremendous speed of scale brands need to keep pace with escalating customer demands, the deluge of data and content, and innumerable customer journey permutations. AI also underpins marketers’ fundamental priorities to make customer engagement more timely, personalized, and results driven.

AI-infused marketing analytics, for example, helps marketers with manual, complex tasks that siphon their valuable time away from more strategic and impactful priorities. AI speeds up important tasks such as integrating data sources, creating segments and rules, designing tests and more.

It also helps marketers automatically discover and generate thousands of discrete segments — instead of hundreds — to evaluate campaign treatments much faster. With AI, marketers can efficiently sort through vast amounts of customer, operational and campaign performance data to optimize offers and personalize campaigns for customers based on their customer-service interactions, in-store behavior, or mobile app use.

The full potential of AI in the enterprise can realized when the unique value of humans and AI technology are harnessed together. AI technology can galvanize humans to create new ideas, challenge their own biases, and reframe their perspectives. Conversely, humans must interpret the outputs and actions of machines in a much broader context — emotional, social, psychological — and use their judgment to guide machines to increasingly perform more relevant analyses.

AI is complementary to the modern marketers’ skills, processes, and technology investments — it will not replace marketers per se.”

At the end of last year, research firm Capterra looked at emerging technologies in SMBs to see how they were reacting to AI and its growing use. It found that AI is often portrayed as a stand-alone technology that can replace human colleagues, a viewpoint that both over and underestimates AI's current capabilities.

As a result, SMB owners avoid investing in AI that could help their human team members work smarter. The survey confirmed this, showing that fewer than one in five business leaders use AI, yet many acknowledge that it is critical.