Four common myths about AI
Artificial intelligence has moved from science fiction to science fact, and the market for AI functionality in business is growing by leaps and bounds. In fact, innovative companies worldwide are mining considerable business value with the implementation of AI capabilities, such as natural language processing, across a broad spectrum of industries. Yet, despite the popularity of AI among early adopters, myths and misunderstandings make it challenging for businesses to integrate and deploy AI effectively.
Let’s take a look at four enduring myths about AI and address why businesses need to reboot their thinking …
1. If you build it, they will come … or will they?
It’s no longer enough to just build and sell a product for the sake of it. The market is crowded and fast-paced as digital startups and legacy enterprises vie for innovative ways to realize their digital future. Forward-looking companies are putting plans in place to generate predictable demand for new sources of digital revenue.
Analyst firm IDC predicts that by 2020, 50% of global 2000 companies will see the majority of their business being dependent on the creation of digitally enhanced products, services and experiences. Pioneering companies that have already made the shift are reporting faster growth than their competition. In this new world, data fuels new and predictable revenue streams on the back of AI, connected things and other digital technologies.
While any company can embark on a digital platform strategy, a differentiated value proposition is essential — and it is required to build at scale. The customer experience is critical. For example, the latency for refreshing images on a virtual reality (VR) headset shouldn’t exceed 10 milliseconds. To accomplish this today, VR gaming headsets require bulky hardware to control sensors and run compute-intensive software.
However, with multi access edge computing (MEC) and AI-based technologies, a mobile edge computing platform would sit close to the point of consumption, taking over the heavy lifting while avoiding unwanted latency. By using MEC and AI, a number of mobile operators, including Deutsche Telekom and Verizon, have recently made announcements related to applications and devices where millisecond response times and customer usage patterns and preferences matter. In other words, by placing AI at the core of their offering and using edge computing, network service providers can make the digital transformation to generate new sources of revenue based on consumer and business demands.
2. One AI size surely fits all, right?
It sounds cliché, but the possibilities with AI are limitless. AI will eventually find its way into just about every product and service. However, not every new product or service will succeed. Success or failure will be based on the business case, and for a number of companies, the right AI use case continues to remain a mystery. In a recent global study conducted by the McKinsey Global Institute, only 20% of C-level executives surveyed said they currently use any AI-related technology at scale, and many firms noted they are uncertain of the business case or the ROI.
Sure, Alexa and Siri seem to have all the answers, but AI for its own sake is not the answer. AI technology is a means to an end, and it should be applied thoughtfully to achieve a real customer value proposition. To achieve revenue growth, operational efficiency and market differentiation, the key question to ask is: What problem are you trying to solve and how can AI play a role?
For example, an industrial equipment manufacturer will benefit from agile machine learning to improve predictions about the onset of failure, allowing maintenance to be scheduled in advance. Likewise, better usage predictions can help a mobile operator reduce the time it takes to resolve network congestion and isolate instability. Failure pattern analysis can enable a software developer to root out priority bugs faster.
Today, an estimated 1,800-plus AI companies have captured more than $16 billion in funding — yet, not all of them will survive. To maximize your product’s value, first and foremost be sure to focus on delivering the best user experience balanced with the best possible business case.
3. AI is revolution not evolution
As digital penetrates every industry segment in varying degrees, there is a sense of urgency for legacy enterprises to go digital fast. Real fast. Even luxury carmaker Porsche is shifting value from horsepower to digital power, with services such as finding nearby parking spots or warning drivers of hazardous road conditions. Soon, it is anticipated that software-based offerings will account for about 10% of the $89,400 sticker price of the new Porsche 911 sports car.
It’s important for companies to map out a digital transformation roadmap — yes, it is evolution — for their mature products before they near the end of life, in order to reduce the negative impacts on the business. The objective is to ensure the core business remains profitable for as long as possible, while successfully incubating fresh product lines shaped by strategic priorities. But be sure to design products around people — the focus should be on how to go above and beyond what the customer expects to meet their digitally focused future expectations. When compared to their peers, revenue growth for companies focused on customer experience outperforms competitors by a wide margin.
Many established companies have already embarked on this digital transformation journey. A prime example is a partnership between GE and AT&T to deploy networked sensors in streetlights. The smart lights can dim or brighten, monitor air quality, keep an eye on parking spots, and even detect and report gunshots. To move digital transformation efforts further and faster, companies need to ensure their R&D organizations have the right business focus, incentives, culture and processes for growth. Ultimately, businesses need to see AI as an evolutionary technology that can deliver outstanding digital experiences for customers.
4. Technology trumps design
Today, the old adage “form follows function” can more accurately be rephrased “form follows emotion.” Designing for the “average user” is outdated. The more personal the experience, the more valuable it will be to the customer — and the more successful your product or service is likely to become. Whether it is a developer console, business productivity application or application for a consumer device, personalization matters.
While AI can be a powerful force for disruption, it is important to focus on what really matters most. From qualitative to end-user participatory research and concept evaluation, the goal of human-centered design is to find new ways to engage with the target audience on their terms. And the stakes couldn’t be any higher. According to a recent study by Salesforce Research, 64% of customers now expect real-time responses and interactions with brands. Half of those surveyed are likely to switch brands if companies do not anticipate their needs.
AI-based machine learning allows companies to collect and analyze data in real-time for a better understanding of the user’s experience with a product. It provides insight and intelligence on how customers are interacting with other users and services. In other words, AI offers the perfect opportunity to create digital growth opportunities, provided that you reimagine products from the end-user’s perspective. This means looking beyond the assumptions and misconceptions of using AI as a technology to using AI-specific use cases with true focus on understanding the needs, preferences, attitudes and motivations of a wide range of customers.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.