Artificial Intelligence Technology Guide

An Enterprise Guide To Modern AI



Steven Lerner
05/29/2019

Artificial Intelligence Technology

Out of all of the solutions leveraged through digital transformation, artificial intelligence technology certainly is in a league of its own. After years of obscurity, AI has now become a common tool in the enterprise, with organizations at least exploring the possibilities of this technology.

In order to understand more about the limits and strategies of AI, here’s a guide about the past, present, and future of this important technology.

What is Artificial Intelligence Technology?

According to Miriam-Webster, artificial intelligence technology refers to the simulation of intelligent behavior in computers. Unlike natural intelligence from living organisms, AI typically describes the types of tasks that computers are able to mimic, such as learning, perception, and communication.

Many enterprises are now leveraging a subset of AI called machine learning, which is when machines use algorithms to perform specific functions without explicit programming, to analyze data. This based on machine learning algorithms that are could be supervised, unsupervised, or reinforced when machines make decisions.

The Evolution Of Artificial Intelligence

The concept of automation and mechanical humans dates backs to ancient myths. As computer science emerged in the late 19th, researchers began to discuss the likelihood of creating an artificial brain. British computer scientist Alan Turing’s research during the 1940s and 1950s laid the foundation for machine learning. The Dartmouth Conference of 1956 is widely considered to be the birth of the modern technology, with researchers agreeing to coin it as “artificial intelligence.”

After a 20-year period full of AI research and discovery, a so-called “AI Winter” commenced between the 1970s to the 1990s, where financial setbacks stymied potential growth. By the 1990s, companies began to invest in AI again, with new tools and algorithms being developed. By the 2010s, deep learning and big data were established, thanks to faster computers.

In 2015, Gartner reported that only 10% of enterprises deployed AI or are were planning to do so. By 2018, AI deployments increased to 25%, and Gartner projects that 37% of enterprises were implementing it in 2019. That’s a 270% increase in just four years.

Industry Experts Explain AI Technology

“AI enables organizations to get a grip on those massive volumes of ever-growing data that are simply outside of the human mind’s capabilities to grasp,” said Mike Leone, senior analyst with ESG. “This enables a new level of productivity that lets people focus on the decision-making based on those insights, as opposed to tinkering with data pipelines, tools, and technology, and ensuring SLAs are being met across all of those things.”

“AI has the ability to predict a chronic condition before it happens in the patient based on their medical history,” said David Chou, technology executive. “That should help with a reduction in readmissions, which is a big theme in healthcare.”

“In the future, we will witness advancements in both AI hardware and software,” said Lian Jye Su, principal analyst with ABI Research. “New edge AI chipsets have powerful edge processing capabilities to enable localized AI functions, such as facial and image recognition, reducing the need to transfer information to the cloud. This means less reliance on cloud, allowing AI to be performed on mobile and remote devices in areas with challenging connectivity. Multimodal AI solutions will combine various AI capabilities, such as image recognition, speech translation, and chatbots, to create a better user experience in customer service, employee training, and management.”

“There is regulatory uncertainty around machine learning and AI, and that organizations will have to act defensively as they plan to scale up their data science efforts,” said Paige Bartley, senior analyst covering data management for 451 Research. “A big part of that will be proper data management practices, of course, but also governance of the end-to-end model development and deployment process. Things like workflow and process management, strong model versioning, and consistent controls for data access permissions are all becoming increasingly important.

AI Trends

Combining AI With IoT

Separately, artificial intelligence (AI) and the Internet of Things (IoT) are impactful. IoT revolves around machine-to-machine communication and devices interacting with the internet. Meanwhile, AI can help companies gain immediate insights from data. Organizations are taking a step further by combining both of these technologies. Known as AIoT, this combination can help an enterprise achieve digital transformation.

There are several advantages to combining AI with IoT. First, this strategy can personalize the user experience, improve customer service, and build deeper relationships with customers. The technology can increase operational efficiency and production, especially by streamlining the hiring process and by spotting potential errors in a supply chain. The combination can improve security and workplace safety by predicting potential risks. Finally, enterprises that are leveraging both solutions are able to reduce costs without sacrificing productivity.

Voice-Activated Technology

Alexa, and other voice-activated assistants are no longer just for the home. In recent years, there has been a trend for this solution, which is powered by AI, to be incorporated into the enterprise. By late 2017, Amazon announced a special workplace version of its voice-activated technology called Alexa for business. Since then, there have been other companies offering similar solutions for the enterprise, including the Google Assistant, Siri (Apple), and Cortana (Microsoft).

Through smart speakers and mobile devices, it is becoming more common to find voice-activated technology in the workplace. The advantages of voice-activated assistants in business include simplifying mundane tasks, analyzing reports, ordering supplies, automatically making phone calls, and setting reminders. Several enterprises are already leveraging the technology and reaping the benefits, including companies in manufacturing, financial services, and education.

Digital Assistants For All Employees

Taking things a step further, a report from Information Services Group suggests that more enterprises will offer AI-enabled personal digital assistants to all employees. Another report from Gartner also said that 25% of all digital workers will have one by 2021. Simply put, there might be a day when every worker in the enterprise has their own digital assistant. This includes text-based assistants in a chatbot, email, or app.

When leveraged correctly, virtual assistants can be a critical component of an enterprise’s plan to automate mundane tasks, improve the employee experience, and achieve digital transformation. Some of the tasks that the virtual assistants could perform include streamlining the supply chain, filtering digital files, scheduling meetings, device support, translations, and transcribing.

AI Challenges And Solutions

Data Management

The success or failure of AI implementation rests with data. Enterprises that are succeeding with AI often have robust data management and centralized data lakes. Failure to conduct proper data management is what sometimes holds an organization back from their AI dreams.

Basic data management optimization is critical, but some organizations fall short of achieving it. Everything from training models to privacy regulations rests on an organization’s ability to manage data. The problem is that some companies lack data management capabilities due to inherent architectural problems. Data is usually siloed in different applications and repositories that are not integrated with each other. This could make it nearly impossible to aggregate and analyze data.

In order to make artificial intelligence technology work, organizations must have a good data science program in place. To achieve this, it is suggested that enterprises conduct a critical review of existing data management and architecture first.

Adopting Voice-Activated Technologies

Although the trend is for more businesses to adopt voice-activated technologies and to reap the many benefits of it, there are several barriers to adoption that must first be properly addressed. Most of the challenges of the technology surround security and authentication. Hackers could easily exploit voice-activated devices/apps and record conversations on them, especially if organizations fail to consider adequate security controls. As a result, some organizations that were planning to use the technology are now taking a step back.

In order to securely implement voice-activated solutions, organizations should ensure that the datasets are personalized, and that security protocols, such as identity verification, are leveraged in this process.

What Will The Future Of Artificial Intelligence Technology Look Like?

Although the current state of AI is focused on augmenting human intelligence, the future could be vastly different, especially as the technology handles more complex tasks. Currently, AI in the enterprise is considered to be ‘weak’ or ‘narrow’ in that its capabilities are limited to specific use cases. The future for AI could allow for more general artificial intelligence (AGI) where the technology would experience consciousness and be able to complete any intellectual task that humans can do.

The idea of machines taking over jobs has been a long-standing fear of workers, but the future is not as bleak as that might sound. The future for AI might be one where organizations embrace ethical AI that improves every aspect of humanity. The technology itself could be very beneficial for enterprises. MIT Sloan Management Review reported that 91% of executives expect new business value from AI during the next five years.

To learn more about the future of AI, and strategies for a successful AI implementation, download our exclusive report, The Evolution Of AI.

RECOMMENDED