See past the hype to ride the GenAI wave in Malaysia

It’s hard to find any news from the IT industry these days that doesn’t contain some reference to GenAI. In less than two years since the commercial debut of large language models (LLMs), this new technology has grown from headline-grabbing interactive chatbots to adding value and intelligence to a range of applications.   

Its evolution has been supercharged by the huge volumes of data on the web that can be used as training material, while the cloud’s computing power is providing the infrastructure to support GenAI’s high resource demands.

According to a Statista report, the size of the GenAI market in Asia is projected to reach $8.45 billion in 2024 and grow at a compound annual growth rate (CAGR) of 46.47% until 2030 to reach $83.42 billion. In Malaysia alone, this market should be worth $179.5 million by the end of this year and increase at a CAGR of 46.5% to reach $1.7 billion by 2030.

Hype versus reality

Statista notes a surge in demand for GenAI-powered chatbots and virtual assistants in Malaysia – applications that can greatly improve the speed and quality of customer service and IT help desks. Using GenAI for data analytics is also attracting more interest.

At the same time, many organizations are experimenting with GenAI but not yet seeing the ROI. They may feel the technology has been overhyped.

Why is this? It may be that they are confining their view of GenAI to tools such as ChatGPT and DALL·E 2, which create new content and images. But this is only one application of the technology. In a business or less consumer-oriented role, however, GenAI can examine massive datasets, make inferences from data and instantly find correlations to support decision-making – or even automate some decisions.

Finding the true business value of GenAI therefore requires thinking about the technology more broadly to understand where it can make a real difference in your organization.

• ALSO READ → Do business intelligently: how to transform your organization with GenAI

Take note of the hype cycle

There are many trends in technology that come and go, but GenAI, like the internet, is a megatrend with lasting power. But where the internet’s development cycle delivered breakthroughs every 18 months or so, major advances in GenAI are arriving at least every six months.

In between, there are smaller GenAI hype cycles around specific applications or use cases. For instance, the spotlight is now on agentics (also called agentic modeling) – the development of autonomous agents that can perform tasks, make decisions and interact with their environment using GenAI.

• ALSO READ → Agentic modeling is shaping the future of GenAI

Meanwhile, in edge computing, small language models (SLMs) are being developed to enable AI at the edge. NTT DATA already has an Edge AI platform that helps our clients speed up the convergence of IT and OT by bringing AI processing to the edge.

Even if your organization is not yet ready to implement the latest GenAI innovations, it’s worth staying up to date with the trends to see how these developments may benefit your business down the line.

Other factors that are worth noting

The belief that GenAI is overhyped may also stem from the high cost and complexity of implementing and maintaining GenAI systems, which require specialized expertise. This puts a strain on budgets well before the technology starts delivering transformative results.

Furthermore, concerns about data privacy and security, regulatory needs and compliance issues could stall the adoption of GenAI. The potential for biased or inaccurate outputs from GenAI models can also undermine trust in the technology.

Integration with existing systems and workflows can also be a hurdle, as organizations may struggle to incorporate GenAI into their operations without disruption.

Follow the guidelines

One way to sidestep the unjustified hype around GenAI is to be as methodical, responsible and strategic as possible when you’re planning to deploy it in your organization.

The Malaysian government has published The National Guidelines on AI Governance & Ethics, an in-depth guide aimed at supporting the development and deployment of AI in the country.

The document recommends steps to take in establishing a data-governance system, training an AI model, and monitoring and learning from both user actions and from the AI system’s performance. It’s also important to perform a comprehensive risk analysis and document your AI governance process in detail.

The guidelines also cover responsibility and ethics in AI, including:

• Keeping AI systems unbiased and inclusive
• Prioritizing the reliability and safety of AI systems
• Data privacy and security
• Transparency of how AI systems work
• Accountability – taking responsibility for the outcomes of AI systems

Organizations that follow the AI governance and ethics process flow set out in the guidelines will quickly benefit from this structured approach and realize the benefits of their experiments with AI and GenAI more rapidly.

How NTT DATA can help in Malaysia

Another way to make GenAI an asset in your organization is to partner with an expert service provider. This gives you access to all the right skills and technology without the need for an upfront investment in your organization.

At NTT DATA, we’re working with organizations in Malaysia to deliver AI and GenAI value. Powered by our ecosystem of partners and our strategy-to-infrastructure capabilities, our solutions help clients get started on their GenAI journey right away. For example:

• Not many organizations can afford to appoint a full-time data scientist, or they may struggle to find the right person in a market where such skills are scarce. NTT DATA’s Data Scientist as a Service aims to deliver all the expertise you need, on demand, through a suite of tailored solutions.
  
  Our service supports your AI efforts with data preparation, data visualization, machine-learning modeling, model deployment, governance and monitoring – everything you need to fast-track your AI journey.
• We also have AI in a Box, a turnkey solution that simplifies AI implementation by combining DataRobot’s enterprise-grade AI platform and Nutanix’s hyperconverged infrastructure – one platform for computing, storage and networking – with Hewlett Packard Enterprise hardware that delivers the performance required by demanding AI workloads.
  
  The technology is complemented by our team of data scientists who have a track record of using DataRobot to build, deploy and streamline models that help our clients achieve measurable business outcomes.
  
  AI in a Box can be implemented in just a few weeks to enable use cases such as the automatic resolution of customer enquiries, document queries and summaries, invoice anomaly detection, data analysis and smart chatbots.

• GenAI initiatives rely on clean, well-governed data to produce accurate outputs. This includes checking data sources checked for bias and making sure that GenAI systems comply with data-privacy and security regulations such as Singapore’s Personal Data Protection Act and the European Union’s General Data Protection Regulation.
  
  We work with Informatica, the data governance and management solution provider, to use their AI-powered, cloud-based data platform to manage your data according to industry best practices that comply with global data-management standards. This leaves your data ready for use in GenAI.

Keep sustainability in mind

We also value sustainability at NTT DATA. The rapidly rising demand for AI and GenAI around the world has led to a spike in energy consumption in data centers and underscored the need for a new technology lifecycle management strategy.

The International Energy Agency reports that electricity consumption from data centers, AI and the cryptocurrency sector could double by 2026, while an IDC report forecasts energy consumption by data centers to grow at a compound annual growth rate of 44.7% to reach 146.2 TWh by 2027 as AI workloads increase.

Our Global Data Centers business is always working to keep our data centers as green as possible. For example, we’re implementing direct chip liquid cooling (circulating a liquid coolant directly over the heat-generating parts of servers), integrating renewable energy sources and using AI to regulate power use and automate temperature adjustments for optimal resource efficiency.

We also rely on infrastructure lifecycle management to make our data centers more sustainable during their lifetime. We recycle or dispose of end-of-life hardware and other ewaste responsibly, in line with the broader adoption of green practices in IT asset management to future-proof organizations, minimize ewaste and resonate with stakeholders and consumers.

These are just some of the ways we help our clients adopt sustainable principles right from the start as they deploy GenAI tools and services in their businesses.

• ALSO READ → Is your organization ready for AI? Here’s what to check first