Unlocking the open AI potential: Ramesh Shanmuganathan on harnessing the power of AI

The advent of ChatGPT is making people nervous about their futures, much like people feared losing their jobs to machines during the first industrial revolution: which many indeed did. However, over time those initial industrial advancements created other jobs, particularly in the services sector, and lifted more people out of poverty. In this interview, Ramesh Shanmuganathan, one of Sri Lanka’s top tech entrepreneurs and corporate leaders, points out that the opportunities and potential to revolutionize businesses far outweigh the risks associated with implementation.

Shanmuganathan, an Executive Vice President and Group CIO at John Keells Holdings (JKH), and Chief Executive of John Keells IT, highlights that one significant advantage of Open AI is personalization. AI can create unique customer experiences by analyzing user preferences, purchase history, and behaviours. This tailored approach boosts customer loyalty and engagement, exceeding user expectations. Operational efficiency is another key benefit. AI-powered chatbots and virtual assistants automate routine tasks, enabling faster processing of requests and enhancing customer satisfaction levels. Predictive analytics, powered by AI, eliminate the guesswork in inventory planning, resource allocation, and assessing emerging product demand. Anticipating customer needs becomes possible, leading to better preparation and service.

Shanmuganathan also posits that AI facilitates enhanced decision-making through comprehensive data analysis. Businesses gain insights into operations and customer behaviour, identifying risks and opportunities in real time. This agility improves responsiveness and ensures better outcomes. Furthermore, AI automation eliminates repetitive tasks, reducing operational costs and freeing up human resources for more valuable activities like customer service and innovation. AI also plays a crucial role in fraud detection, promptly identifying fraudulent activities and protecting data integrity.

Despite these benefits, challenges exist. Shanmuganathan says data quality, privacy, and security are critical concerns. Organizations must ensure accurate and secure data collection, processing, and analysis. Bias and transparency issues need to be addressed, guaranteeing fair and unbiased AI systems and building trust with customers. The adoption of AI requires extensive training and upskilling of staff. Organizations must invest in training current employees and recruiting new talent proficient in AI development and implementation. Strict regulatory compliance in sectors like healthcare and finance adds additional hurdles to AI adoption. By developing effective strategies to tackle these challenges, organizations can harness the full potential of AI while ensuring responsible and beneficial AI adoption in the services sector, Shanmuganathan explains in this interview.

ChatGPT has made AI more accessible and interesting. What do these new developments with AI in search and similar areas mean for businesses now?

Open AI initiative has made AI accessible to more people than ever before. The continuous improvements brought about in the successive releases such as ChatGPT-4 keep unravelling new areas to explore.

The new developments with AI in search and similar areas provide a great opportunity for businesses to improve customer experience, and business operations, and enhance their overall performance.

Some of the key use cases where businesses can leverage these developments are:

1. Enhance Customer Experience – provide more tailored and helpful information and solutions to their customers based on their preferences and behaviours.

2. Increase Operational Efficiency – automate routine tasks and processes and optimize their workflows and systems.

3. Enhance Decision Making – leverage data-driven insights and predictive analytics to anticipate future trends and opportunities and mitigate risks.

4. Improve Search Results – more accurate and relevant search results that can significantly enhance their customers’ online experience. This can lead to increased engagement, higher conversion rates, and improved customer retention.

5. Personalize Recommendations – to provide personalized product recommendations which can lead to increased sales and customer loyalty.

6. Better Customer Service – handle customer inquiries and support requests more efficiently and accurately than human agents, providing faster response times and reducing human error.

7. Predictive Analytics – helps businesses anticipate future trends and identify emerging opportunities. By analyzing past data patterns, businesses can make informed decisions on products/services to offer, promotions to run, or inventory to stock.

8. Robotic Process Automation – streamline repetitive tasks such as data entry and processing, freeing up employees’ time to focus on more critical tasks.

However, we must be conscious of the fact that these new developments also pose some challenges for businesses, such as:

1. Data Privacy and Security – as they need to ensure that the data they collect, store and process via AI systems are protected from cyberattacks and comply with relevant regulations.

2. Adaptability and Integration – as they need to invest in the appropriate infrastructure, talent and training programs to incorporate AI into their business operations and culture.

3. Ethical Issues – as they need to consider and address potential ethical concerns related to AI’s impact on people’s livelihoods, jobs, rights and values.

Therefore, these new developments with AI in search and similar areas provide a gamut of opportunities to improve business operations, enhance customer satisfaction and optimize overall performance. These developments are expected to revolutionize the world of business in the coming years by enabling more efficient workflows, better decision-making and increased customer satisfaction, leading to greater profitability and success. Businesses should embrace innovation and adopt AI-powered solutions strategically while being mindful of the potential benefits and challenges that they entail. By doing so, they can gain a competitive edge in their respective industries and achieve better outcomes for their customers and stakeholders.

What is generative AI and how is its utility different from what we have come to expect from AI?

Generative AI is a type of artificial intelligence system that can produce various types of content, such as text, images, audio and synthetic data, by learning from existing data and mimicking its patterns and structure. It’s a subtype of artificial intelligence that involves teaching machines to learn patterns and generate original content such as images, text, video, and audio. Unlike other forms of AI, which typically focus on performing predefined tasks, generative AI utilizes deep learning models to generate entirely new content based on the input data it has been trained on. Some of the examples are,

• ChatGPT, a chatbot that can generate coherent and engaging text based on user input.
• DALL-E, an image generator that can create realistic and surreal images based on natural language prompts
• Midjourney, an art generator that can create high-quality images based on natural language prompts

Generative AI is different from what we have come to expect from AI because it enables machines to create something entirely new and unique. While traditional AI systems are designed to perform specific and specialized tasks like recognizing objects in an image or translating languages, generative AI can create new creative works, improvising and learning as they go along. The utility of generative AI is different from what we have come to expect from AI in several ways:

• Generative AI can create new and original content, rather than just analyzing or classifying existing data. This can enable more creativity and innovation across various domains and industries.
• Generative AI can perform multiple tasks across different modalities, such as text, image, and audio, rather than being limited to a specific domain or task. This can enable more flexibility and versatility for users and applications
• Generative AI can produce high-quality and realistic content, as well as surreal and novel content, depending on the user’s preference and purpose. This can enable more personalization and diversity for users and applications

However, generative AI also poses some challenges and risks, such as:

• Data quality and quantity, as generative AI systems, require large amounts of high-quality data to train on and generate accurate and relevant content. This can be costly and time-consuming to obtain and process.
• Data privacy and security, as generative AI systems may expose sensitive or personal information from the data they use or generate. This can raise ethical and legal concerns for users and applications
• Data authenticity and trustworthiness, as generative AI systems may produce fake or misleading content that can deceive or manipulate people. This can have negative social and political implications for users and applications.

Generative AI has several real-world applications, including art and music creation, product design, and even video game design. For example, a generative AI system can be trained on existing patterns and styles of music to compose an entirely new piece of music that sounds like it was created by human composers. Similarly, companies can use generative AI to create new designs for products, such as clothing or furniture, based on the machine’s analysis of existing products and customer preferences.

Although generative AI is still in its early stages of development, it has huge potential in many industries. As this technology advances further and becomes more accessible, we may see entirely new business models emerge that rely heavily on the creativity and ingenuity of machines. The potential applications of generative AI are limitless, and it will undoubtedly revolutionize the industry in ways we cannot yet imagine, but it should be used responsibly and ethically, with proper safeguards and regulations in place to ensure its quality, security and authenticity. By doing so, generative AI can offer many benefits and opportunities for users and applications across various domains and industries.

How might AI help drive value along the supply chain, in businesses with enough scale?

AI has the potential to drive value due to its inherent ability to process massive amounts of data on a real-time basis and as such brings along the following advantages to any business or business process:

1. Real-Time Data Processing – capable of processing massive amounts of data in real-time, which makes them essential for supply chain management that involves numerous touchpoints and interactions.

2. Decision Making – analyze data from various sources to help businesses make strategic decisions related to supply chain operations, including sourcing raw materials, production scheduling, and distribution.

3. Risk Management – identify potential risks in the supply chain, such as product shortages, transportation disruptions or supplier issues. By recognizing those risks in advance, companies can take measures to mitigate unforeseen events.

4. Customization – the optimal combination of delivery times, inventory stock levels, transportation modes, and routes for individual customers based on their preferences.

5. Supply Chain Visibility – monitors the location, quality, and condition of products during transit- this provides deeper insight into the supply chain and facilitates more accurate predictions on lead times and delivery schedules

The key value drivers of using AI in the supply chain would be:

1.  Improving demand forecasting and production planning, by using AI to analyze historical and real-time data from various sources and generate accurate and timely predictions of customer demand and optimal production levels.

2. Enhancing inventory and warehouse management, by using AI to optimize inventory levels and locations, reduce stockouts and excess inventory, automate warehouse operations and improve order fulfillment.

3. Streamlining logistics and transportation, by using AI to optimize routing and scheduling, monitor and track shipments, reduce fuel consumption and emissions, automate driverless vehicles and drones, and improve customer service.

4. Increasing supply chain resilience and agility, by using AI to model and simulate various scenarios, identify and mitigate risks, respond to disruptions and changes in demand or supply, and enable collaboration and coordination across the supply chain network.

Some of the well-established use cases examples of businesses that are using AI to drive value along the supply chain are:

• Walmart uses AI to forecast demand, optimize inventory levels, automate warehouse operations, improve delivery speed and accuracy and reduce waste.
• Unilever uses AI to plan production, optimize transportation routes, monitor supplier performance, reduce carbon footprint, and improve customer satisfaction.
• UPS uses AI to optimize delivery routes, track shipments, predict delivery times, automate driverless trucks, and enhance customer experience.

By using AI to drive value along the supply chain, businesses with enough scale can achieve higher efficiency, lower costs, better quality, greater customer satisfaction, and more sustainability. Further, businesses with enough scale can leverage AI’s benefits to optimize their supply chains by analyzing vast amounts of data, improving decision-making processes, mitigating risks, customizing offerings to customers based on their preferences, improving visibility throughout the supply chain, and maximizing operational efficiency. While implementing AI-powered solutions requires significant investment in software and hardware infrastructure and skilled personnel, the long-term benefits can be tremendous, especially when the company operates across multiple regions.

What are the opportunities and challenges for AI in the services sector?

The services sector, which encompasses a wide range of industries such as hospitality, healthcare, and finance, presents several opportunities for AI to enhance efficiency, quality, and customer experiences. Some of the compelling ones are,

• Personalization – One of the significant benefits that AI brings to the services sector is creating unique customer experiences based on user preferences, purchase history and past behaviours. Boost customer loyalty and engagement by tailoring services that meet or exceed user expectations.
• Operational Efficiency – With AI-powered chatbots and virtual assistants, service-based businesses can automate routine tasks and process requests faster resulting in enhanced customer satisfaction levels.
• Predictive Analytics – Leveraging AI-powered predictive analytics eliminates the guesswork in planning inventory levels, resource allocation, and assessing demand for emerging products or services. This enables you to anticipate customer needs and prepare accordingly.
• Enhanced decision-making: By leveraging large amounts of data, AI can help businesses make informed decisions based on a comprehensive understanding of their operations and customers. AI-powered tools can also help identify risks and opportunities in real-time, improving the agility and responsiveness of businesses.
• Improved efficiency and cost savings: Automation through AI can eliminate repetitive tasks that take up significant human labour, reducing operational costs and enabling staff to focus on higher-value tasks such as customer service and innovation.
• Fraud Detection – AI has improved the ability of service providers to detect fraudulent activities promptly and accurately during transactions while identifying security threats, and protecting users’ data integrity, reputation and trust

While adopting AI presents significant opportunities for cost savings, operational efficiency, personalization and improved customer satisfaction levels in service industries. Successful AI implementation will require businesses to address challenges such as data security, workforce adaptation and ethical considerations and challenges associated with them such as

• Data quality and privacy: One of the main challenges in implementing AI in the services sector is ensuring the quality and accuracy of data used to train algorithms. Additionally, collecting and processing sensitive customer data raises concerns about privacy and security.
• Bias and transparency: Another challenge is ensuring that AI systems operate without bias or prejudice. This requires careful consideration of how data is collected, processed, and analyzed to ensure that algorithms are fair and unbiased. Further, there is a need for transparency in how AI decisions are made to build trust with customers.
• Training and upskilling: Adopting AI in the services sector requires extensive training and upskilling of staff, which can be a significant investment of time and resources. Organizations must not only train current employees but also recruit new talent with skills in AI development and implementation.
• Regulatory compliance: In industries such as healthcare and finance, AI solutions must comply with strict regulations around data privacy and security, which can create additional barriers to adoption.
• Customer acceptance: Finally, customer acceptance of AI-powered services can be a barrier to adoption. Some customers may feel uncomfortable interacting with chatbots or automated systems and prefer human interactions, while others may have concerns about the use of their data.
• Ethical and social issues – AI systems may impact human values, rights, and responsibilities, such as fairness, accountability, transparency, and trust. This can have legal and moral implications for service providers and customers
• Human-AI interaction and collaboration – as AI systems may need to communicate and cooperate with human users and workers effectively and appropriately. This can require designing user-friendly interfaces, ensuring explainability and feedback mechanisms, and providing training and support.
• Adaptability & Integration – AI provides solutions that require major changes in workflows and processes. Shifting to new AI-enabled software and technology can be challenging for employees who may need retraining and adaptation to embrace these changes seamlessly.

Overall, while the benefits of AI in the services sector are clear, organizations need to consider these challenges and develop strategies to address them effectively. This includes not only technical considerations but also ethical and social implications of AI adoption.

If organizations have large amounts of data, how should companies navigate for generative AI and AI broadly to build a competitive advantage?

If a company has a large amount of data, using AI and generative models can be a powerful tool to gain a competitive advantage. Some of the key strategies they could adopt to build a competitive edge are:

• Focus on specific use cases: AI is most effective when it is applied to specific use cases that align with the business goals and strategic objectives of the organization. Companies should identify the activities where they believe AI can generate the most significant benefits and match these with the data that they have available.
• Develop strong AI infrastructure: Building robust data architectures and developing sophisticated algorithms are key components in leveraging AI successfully. An effective AI infrastructure is crucial for analyzing large quantities of data, extracting insights, and designing predictive models
• Develop strong AI infrastructure: Building robust data architectures and developing sophisticated algorithms are key components in leveraging AI successfully. An effective AI infrastructure is crucial for analyzing large quantities of data, extracting insights, and designing predictive models
• Data Collection: To navigate generative and AI technologies effectively, businesses need a robust data strategy that covers the entire organization. This may involve breaking down data silos and investing in technologies that help collect, clean, and store data efficiently. In some cases, companies may need to leverage third-party data sources to enrich their existing data sets further.
• Selecting the right AI algorithms: Not all AI algorithms are created equal. Some machine learning models are better suited to certain tasks than others. Companies should identify the use cases where AI can provide the most significant impact and then select algorithms that are best suited to those specific scenarios. Additionally, organizations should set performance expectations aligned with relevant metrics to track outcomes accurately.
• Automating processes: One of the primary benefits of AI is process automation. By automating repetitive, time-consuming tasks, businesses can free up their employees to focus on more value-adding activities. However, companies should approach automation strategically so as not to replace human workers entirely. Instead, they should aim to augment human operations with AI tools.
• Constantly Improve Outcomes: Deploying an AI application is just the beginning of the journey. Once an application is in place, businesses must continually monitor its performance and adjust its inputs and outputs accordingly. This may require fine-tuning algorithms, adjusting data sets, or revising performance metrics.
• Foster Collaboration: Effective use of AI requires cross-functional collaboration between business analysts, data scientists, software engineers, and other professionals from different areas of expertise. By working together, the various stakeholders can ensure that everyone understands the interdependent tasks required for success.
• Hire the right talent: Building an effective AI team requires a diverse range of skills in domains like ethics, data architecture, software development, and machine learning. Companies should invest in recruiting top talent from a variety of backgrounds and disciplines to create an effective team.
• Continuously iterate and improve: Ultimately, the goal of using AI is to continuously improve processes and deliver better outcomes for customers. Companies should closely monitor performance metrics and continuously refine their models to achieve greater efficiency, effectiveness or accuracy

Overall, for companies with rich sources of data, properly implemented AI and generative models can offer significant opportunities for building a competitive advantage. By focusing closely on specific use cases aligned with their overall business goals, building strong infrastructure, adhering to ethical standards, hiring top talent, and steadily iterating towards optimal outcomes a company can successfully leverage AI to gain a competitive advantage. However, it’s essential to remember that AI adoption is a journey, not a destination. Companies that succeed in this area will be those that remain flexible and continuously adapt to changing circumstances.