Why Companies Still Struggle To Incorporate AI Into Existing Business Models

• Read original article here

Artificial Intelligence (AI) is impacting the world in different ways. From core industries such as manufacturing, all the way to healthcare, AI and deep learning models are changing processes and making things more efficient.  

And despite turbulent times, a report pinpoints that nearly half of the companies have accelerated their AI strategies, 20% doing so “significantly,” betting their AI projects will have a positive impact on their organisation’s resiliency, efficiency, and innovation. Yet almost half (49%) of respondents feel their company is behind in their AI journey, suggesting a critical gap exists between the strategic need and the ability to execute among business leaders and technologists. 

They say there is still a lack of knowledge and clarity on the applications of the technology that has been a roadblock due to which AI adoption has been slower than anticipated in the enterprise. A large company jumping onto the AI bandwagon without a strategy can cause a painful journey and disappointment. Technology is an integral part of today’s business and AI is an important technology.

According to many analysts, there is a lack of knowledge and clarity on the applications of artificial intelligence in organisations. Regardless of the fact that it has taken decades for the algorithms to come to life, neural networks theorised decades ago have created a paradigm-shifting effect across different business sectors. These AI-based engines are powering innovation including natural language processing, computer vision, facial and speech recognition and robotics.

“Far too many people think of AI as an abstract force in the hand of expert geeks. Instead, AI is there for all of us to use. The bottleneck of widespread application of AI is not technical expertise, but the diverse set of people developing an intricate understanding of AI and putting it to work,” says Philipp Gerbert, well known AI analyst in a talk.

AI is increasingly being utilised by software vendors and AI solution providers as embedded products and services to give more value over a host of business obstacles. This journey has origins in core enterprise applications like ERP and CRM, but now, almost all sectors are using AI in their supplemental processes such as customer support, recruitment sales, or marketing. For example, a bulk of banks and insurance companies in India have AI-powered chatbots, which are quickly becoming the first point of customer interplays. 

For retail companies, AI creates an opportunity to connect the gap between virtual and physical sales channels. From daily task management to getting customer insights, AI is a fundamental technology in a business setting such as retail. Factors such as demand for supply chain optimisation, improved decision making of business leaders, and forecasting among retailers is making AI an integral part of the business. 

Cutting-edge companies are already finding patterns in user behaviour that can lead to exceptional products or features in existing products, which is giving them an extreme advantage over other businesses. Take computer vision (CV) for example. With computer vision, we can create a system that does a subset of things that the human visual system can do. In CV, a system can analyse a picture taken by a camera and understand what’s in the picture. For example, it can recognise objects like cars, streetlights and of course people.

Computers can perform object recognition through a network of nodes called neural networks. An image can be fed into the network, and convolution happens at these nodes. This kind of technology can be used for various business scenarios and lead to incredible amounts of productivity and efficiency. For example, you can leverage computer vision-based licence plate recognition to run an automated car parking business. Of course, the information from registration, billing and computer vision-based license plate recognition systems would have to be integrated to automate the entire process. 

Most of the recent innovation in AI happened with increasing processing power in data. Also in the last ten years, two of the longest standing problems of AI have been somewhat cracked, i.e vision and language, both of which propelled AI into the real world. Vision is needed for robots, drones and self-driving cars and language is crucial for interacting with humans, accessing documents and knowledge. While the two technologies are not perfect yet, they are making a massive impact already. 

“So far, the focus on Enterprise AI has leveraged standard models published by researchers in computer vision, speech, and NLP. These have made various new products possible and simplified consumer experience. As the field matures, capabilities like differentiable programming are making it possible to use AI technologies to solve core business problems. Combining the power of new hardware with flexible programming stacks and programming languages, it will become possible to embed business logic in Enterprise AI systems,” told Viral B. Shah – Co-Founder and CEO – Julia Computing during an interaction with Analytics India Magazine.

As companies embark on integrating artificial intelligence, it might be useful to consider both an AI framework and an AI strategy. When you think about a company, you need to think about multiple dimensions such as the AI platform, tools to support the development, the AI/data science team, the business processes where AI will be integrated, business capabilities that have to be enhanced, organisation structural changes, and external partners among many other dimensions. 

This framework can be visualised in many ways. Along each dimension, the business may have to make some decisions, which form the AI strategy such as using an on-premise AI environment or using a cloud provider. To make this work, companies need to spend a lot of time figuring out various skills, time and financial resources. The best way to do it to ultimately align with the core business strategy.

Just like how a business needs a strategy, AI does too. But both are tightly linked. Think of today’s cloud kitchen — the business model would not have been possible (given tight profit margins) without the mobile devices, and there is no way cloud kitchen would have maintained profitability without integrating AI directly into their business models. The same applies to aggregator cab services like Uber, streaming services such as Netflix or Spotify, and house rental services like Airbnb. AI has become the main factor in the most advanced business models.