AI Disruption is unsettling, but it can also serve as a catalyst for innovation and transformation.
MEDIA 7: Could you please tell us something about your career journey so far, and what made you want this position at IDC?
RITU JYOTI: Here's the link to my bio on idc.com – this should give you a good snapshot of my career so far: Ritu Jyoti, GVP, AI and Automation Market Research and Advisory Services. AI is one of the most disruptive innovation of our lifetime. The opportunity to lead a group of brilliant minds and be responsible for IDC's thought leadership in this space, while changing the way the world thinks about the impact of AI on business and society got me excited about this position. It has empowered me to make tangible impact on technology buyers and suppliers business outcomes.
M7: You started your journey with IDC in 2017. Where was the company at when you started, and where is it now? What changes have you seen take place for the better when it comes to smart technologies like AI?
RJ: Although AI has been around since the 1950s, 2020 was the year that strengthened the value of enterprise AI. In Ernest Hemingway’s novel "The Sun Also Rises" you will find a dialogue between two characters which goes like this: “How did you go bankrupt?” Bill asked. “Two ways,” Mike said. “Gradually and then suddenly. " "Gradually and then suddenly" might be one of the most profound insights into how disruption happens today: First you cannot see nor feel it. It is a profound lesson for everyone who is in a legacy business (and make no mistake — most businesses can be considered legacy businesses these days) and a good motivator for those who are ahead of the curve. The first smartphones were undoubtedly impressive, but they did not kick a massive dent into the universe — until the iPhone and later Android phones came out and suddenly shifted the equation.
The food and beverage industry’s shift to online was similar — gradual for the last few years, then suddenly because of the global COVID-19 pandemic. In fact, at PepsiCo, they have estimated that the pandemic has accelerated the adoption of online grocery by 3-5 years. Disruption is unsettling, but it can also serve as a catalyst for innovation and transformation. We have now entered the domain of AI-augmented work and decisions across all the functional areas of a business. Responsible creation and use of AI solutions that can sense, predict, respond, and adapt at speed is an important business imperative. AI, machine learning, and natural language processing are beginning to play a much larger role in enterprise businesses, whether it is in customer service, customer relationship management, or even learning initiatives.
Read More: Gartner’s Akhil Gopinath advises, ‘To pivot your business model and grow out of the pandemic phase’
A hybrid workforce, virtual recruitment, and a heightened focus on diversity and inclusion have introduced new dynamics and intensified existing ones.
M7: That is interesting. Where do you think we are going from here into the future with these smart technologies?
RJ: In 2020, numerous enterprises discovered just how much AI and ML tools could help their organization remain stable and even continue to grow despite the turmoil rolling through the markets. While AI is still primarily utilized for IT, cybersecurity, and engineering and production, it's seeping into more business-critical functions such as marketing, legal, HR, procurement, and logistics. The cost and availability of sophisticated machine learning models are providing organizations with the ability to monitor, analyse, control, and customize machine-led processes across a wide variety of horizontal business functions. For example, AI can be used in procurement for spend analysis to contract management and strategic sourcing. Similarly, AI in supply chain and logistics provides real-time tracking mechanisms to gain timely insights including the optimal times by where, when, and how deliveries must and should be made. Such powerful multidimensional data analysis further aids in reducing unplanned fleet downtime, optimizing fuel efficiencies, and detecting and avoiding bottlenecks. It provides fleet managers with the intelligent armour to battle against the otherwise unrelenting fleet management issues that occur daily.
In the world of digital marketing, AI can streamline and optimize marketing campaigns. It can also eliminate the risk of human error. AI has the potential to both curate and generates content, then place it in front of the right people on the right platforms. On the strategy side, AI has the potential to help marketers map out an end-to-end content strategy. Some marketing tools are already providing this feature and are expected to generate comprehensive reporting on content initiatives, with little to no human labour involved. Used responsibly, AI is fuelling HR's transition from administrative to strategic to mission-critical as we face a very different landscape than we did 12 months ago. A hybrid workforce, virtual recruitment, and a heightened focus on diversity and inclusion have introduced new dynamics and intensified existing ones. Increasingly, businesses will embrace AI-powered HR platforms to stay ahead.
The computer vision AI field has come a long way since the first forays in the 1960s to detect object edges and categorize simple shapes. Powered by advances in graphical processor units, modern neural networks have seen object detection accuracy rates go from less than 40% to more than 90% in the past decade. This is supporting an explosion of use cases for computer vision AI. Computer vision has the potential to bring in some real value in retail, healthcare, financial services, manufacturing, and other industries. It is also poised to transform customer, field, and IT service management, along with marketing promotion, supply chain management, and procurement functions.
For example, in retail, there are more than a dozen good use cases for computer vision including automated checkout, loss prevention, risk reduction, queue management, inventory management, and customer traffic monitoring, and some of these require immediate accurate results. Neural networks used for computer vision applications are easier to train than ever before but that requires a lot of high-quality data. This means that the algorithms need a lot of data that is specifically related to the project to produce good results. Even though images are available online in bigger quantities than ever, the solution to many real-world problems calls for high-quality labelled training data. That can get rather expensive because the labelling is typically done by a human being. Data augmentation and production of synthetic data to help build the adequate size of the training data set is still evolving. In addition, there is a labour shortage in data service candidates, roughly 140,000 in 2021. To alleviate these challenges, the deployment of low-code/no-code computer vision environments will drive companies to use subject matter experts to train computer vision models instead of data scientist teams. These teams will leverage pre-trained models, transfer learning, and synthetic data generation to create novel computer vision solutions for companies at a fraction of the historic cost and in rapid time.
The increased frequency and severity of natural catastrophes is making climate change a strategic priority for organizations worldwide. Adverse weather events can and increasingly will wreak havoc on business' core operations: disrupting supply chains, forcing mass evacuations due to wildfires, flooding facilities near coasts, halting outdoor activity due to extreme heat, and making certain regions less habitable. For instance, a bank might want to understand the expected impact of increased hurricane activity on property damage along the coast as it assesses its real estate loan portfolio. A government may seek to make targeted investments to bolster its country's critical infrastructure in the face of more punishing weather conditions. An international hotel chain may find it worthwhile to better understand long-term weather patterns. More than 8,000 suppliers of goods and services to large corporations reported that $1.26 trillion of their revenue is likely at risk over the next five years because of climate change, deforestation, and water insecurity, according to a report by CDP, a non-profit platform for corporate environmental disclosures.
Weather and climate forecasting services have existed for decades, but historic climate data is proving to be less effective in helping predict future events. We are seeing a growing industry of climate-risk intelligence companies that combine troves of historical and current data with sophisticated artificial intelligence techniques and systems, including neural networks, to help account for the increasing velocity of events linked to climate change. Neural networks have the advantage of being able to tackle unstructured information, such as weather and climate maps or graphs of historical temperature data, more easily, contributing to more refined and accurate forecasts. Neural networks also can learn to weigh the latest data more heavily in their calculations. Neural networks factor in time, allowing older information to decay in importance. The world's ice, for example, is melting at an accelerated pace. A neural network will recognize that and factor out older ice melt rates. But the technology does have limitations. Companies need adequate data to train their models, and there isn't always enough data. One example is hail, where limited observations make it hard to train AI models. We are predicting that "By 2027, 50% of the G2000 Organizations Will Invest in Neural Networks–Powered Climate Hazard Assessment, Adaptation, and Identification of Opportunities, Driving 25% Profit Growth."
M7: Consumer behavior has witnessed a whirlwind of changes in the past few years, do you think it is the best opportunity for enterprises to leverage Conversational AI?
RJ: IDC has identified that, over the past 12–18 months, more and more organizations are deploying conversational AI applications into production. Most of these conversational AI applications are focused on customer service and customer care, but an increasing number are found in IT support, human resources, and even sales and marketing. One of the primary impetuses for this increase has been the tremendous improvements in natural language processing, natural language understanding, speech recognition (speech to text), and natural language generation that have occurred over the past three years. In the past, these technologies were dependent on heuristics and algorithms developed by humans (i.e., conversational linguists). However, as machine learning and deep learning capabilities have improved, utilizing more powerful GPUs and CPUs as well as the massive amounts of conversational data from recordings, videos, and even social media applications, these technologies are now increasingly driven by deep learning models. In addition, research work done by several commercial vendors and academic R&D centres has resulted in new open-source advanced language models that process and "understand" both written and spoken language better than ever before.
Built by OpenAI, an independent AI research and deployment company with investment and association with Microsoft, GPT-3 is a massive natural language model driven by deep learning to provide a better understanding of the everyday language spoken or written in the world today. Google has developed BERT, a transformer-based machine learning model for natural language processing that specializes in providing contextual clues about words and phrases. IDC believes that more and more conversational AI applications will be based on these types of deep learning models. The usage of these models will ensure that conversational AI applications act and sound more realistic — understand and respond to users' queries more accurately.
Essentially, AI, machine learning, and NLP are changing the face of brands around the globe. With AI-enhanced chatbots, businesses can enhance the level of speed, satisfaction, and personalization when customers interact with brands through all of their channels. By using behavioural and emotional prediction, brands are better able to fully understand what their customers are going through on their journey. Using AI-enhanced data extraction and analysis, brands can more effectively leverage all of the consolidated data. Using NLP, business applications can understand the various forms of communication in a brand’s arsenal, facilitating true digital transformation. AI is here to stay, and brands should be using it to play larger and greater roles in their DX strategy.
Read More: Bidnamic's CMO, Alun Davies, says that marketers are now more like software developers
AI is here to stay, and brands should be using it to play larger and greater roles in their DX strategy.
M7: That’s true. AI is definitely here to stay. How important do you think it is to leverage technology investments to improve business agility?
RJ: Technology investments like AI are critical enablers to business agility. There are numerous use cases of AI that show a tangible impact on agility. The technology’s unique ability to provide critical insights delivers efficiencies in many layers of company operations, from decision-making to customer service to product design, in turn increasing productivity. AI’s predictive analytics skills combined with immense computing power significantly streamline and accelerate the speed and accuracy of decision-making. In some cases, this involves quickly capturing, analysing, and presenting the relevant data, along with a recommendation on the next steps. It can even go as far as making decisions itself, something we will begin to see more of as the technology continually matures. As machines possess the ability to process inputs mathematically with unprecedented speed and make decisions based on previously accumulated data, the resulting outputs are highly accurate, eliminating the risk of human error.
This speed of accurate decision-making leads to quicker resolutions, enabling the agile business to identify and fix problems before they even begin to develop. Ultimately, humans will never be able to work 24/7. AI can work continuously without breaks or dips in productivity, consistently providing the same quality of service. In this way, it can help organizations reach levels of agility that would have been previously unattainable without the use of technology. Through smarter decisions, quicker resolutions, reduced costs and minimal errors, it is clear how AI can improve agility. But there are steps organizations should take to effectively implement this technology. In some ways, AI and agility form part of a lifecycle: businesses need to be agile to introduce disruptive tech like AI, whilst AI in turn can help businesses achieve true agility. Laying out the initial groundwork will prove invaluable in embarking on the journey of adoption. Organizations should first identify and agree on the problem to be solved. Every business has areas that could be improved or processes that would benefit from becoming more productive.
Establishing these inefficiencies will allow organizations to better determine AI’s use cases and see the long-term advantages. Businesses will also need to understand their internal capabilities, looking at their existing people, technology, and data. Implementing new tech does not necessarily require a complete overhaul. Many modern solutions slot cohesively into the existing technology stack, enhancing existing processes and providing new capabilities on top. When implementing AI, starting small enables organizations to prove the model and the outcomes before expanding to other areas of the business. The agility journey is a marathon, not a sprint. Successful digital transformation is an important step, but improving culture, structure, skills, employee experience, and attracting and retaining talent is equally as vital. Whilst AI can quickly deploy functional outcomes, other technologies can also help to improve business agility, such as Robotic Process Automation (RPA), Business Process Management (BPM), Orchestration and Analytics. Organizations should consider all available options and how they can work in tandem, complementing each other for the most suitable, tailor-made approach for the business. With technology, there is no one size fits all, and a combination of digital capabilities can transform the overall experience.
Whilst organizations cannot solely rely on AI to help them become agile, businesses can use the tech to resolve many of the pain points currently holding them back. Its data insights drive better decision-making, it can fix issues before they escalate, and it reduces costs, all while maintaining consistent and continuous service delivery. Technology, people, culture and structure form the essence of the ‘Operating System of the Enterprise’. Those businesses who maximize the potential of each pillar will discover and accelerate their route to achieving agility, reaping the benefits as a result.
M7: Last but not the least. What is your go-to resource – websites, newsletters, any other – that helps you stay updated with the revolutions happening in the digital space?
RJ: WSJ.com AI section, HBR and MIT Sloan Review AI newsletters, Industry conferences, podcasts, webinars, Vendor newsletters, IDG Think Tanks, towardsdatascience.com etc.