Artificial intelligence (AI) is rapidly ascending as a transformative force in the business world. Traditionally, AI technology has found its home in digitally transformed businesses that have successfully integrated cloud technology. However, a notable shift is occurring, with a growing demographic of relatively inexperienced, non-tech-inclined businesses and consumers eager to explore AI possibilities. According to IBM research, this evolution is underscored by the fact that 44 percent of organizations are actively working on integrating AI into their current applications and workflows, emphasizing the growing importance of making AI accessible to a broader audience.

As the appeal of AI lies in its advantages, including innovative automation capabilities, enhanced user-friendliness, and the potential for higher returns on investment, a recent survey has further cemented the importance of AI, indicating that an AI-first strategy has emerged as a key indicator of AI maturity and is directly linked to achieving superior returns on investment.

In Singapore, the AI deployment rate currently stands at 39 percent, signifying a growing interest in leveraging AI technologies to enhance business operations. However, it is vital to note that 46 percent of businesses in Singapore are still in the exploratory phase, indicating the existence of challenges hindering full-scale AI adoption, despite the promise of exponential growth and increased ROI.

AI transformations shaping industries

Massive developments in large language models have propelled AI technologies into the realm of unsupervised learning. An illustrative example is ChatGPT-4.0, showcasing exceptional accuracy, adept data comprehension, and self-correction capabilities, surpassing the performance of its predecessors. Additionally, enhanced large language models have also elevated sentiment analysis and summarization capabilities, even exceeding human levels of accuracy and efficiency.

These advancements have not only reduced the cost and effort required for AI development but have also made AI technologies more accessible to a broader range of users. Moreover, improvements in AI’s natural language understanding – with a ~15 percentage point increase – have led to enhanced accuracy in various natural language processing (NLP) tasks. Additionally, AI-facilitated software engineering transformations are gaining ground. Tools like GitHub Copilot and Amazon Code Whisperer are now assisting developers in their coding tasks, accelerating development, and making it more accessible. Overall, these developments bring about substantial benefits, from increased operational efficiency for all businesses to reduced expenses and lead times for traditionally costly projects.

Hurdles in implementing AI

Although the potential of AI is considerable, organizations still encounter challenges in the implementation and integration of AI solutions into their business operations. One significant obstacle is the lack of AI use knowledge – seeing that new and potential AI adopters may wrongly perceive AI integration as a universal, one-size-fits-all solution. However, AI technologies are inherently challenge-specific solutions, and users must consider their operational issues and desired outcomes when identifying the right AI solution. Furthermore, each AI technology serves a specific purpose, whether it is analyzing tenders in renewable energy or enhancing video streams by removing occlusions.

Another critical barrier that hinders widespread AI adoption, despite the acknowledgment of its potential benefits, is the entry costs associated with accessing advanced AI models and the necessary hardware infrastructure for training. While some AI models are open source, substantial computational resources are often required for training, which may not be readily available. Hence, the most affordable and easiest way to deploy LLM’s at this point is to use them in the Cloud provided by AWS, Microsoft, and GCP. In addition, the complexities associated with implementing AI systems can be challenging as well, particularly for small and medium-sized enterprises (SMEs).

Data Privacy and Compliance constitute yet another challenge. AI often involves the processing of sensitive data, necessitating organizations to prioritize data privacy and compliance. Adherence to relevant regulations is critical for building trust and credibility in AI adoption. Furthermore, navigating the complex web of data privacy laws and regulations can slow down AI adoption, as companies prioritize safeguarding sensitive information.

The key to success: Focus on core business value

AI is a transformative force that can introduce automation and create new revenue streams through innovation. However, according to Gartner, up to 85 percent of AI projects fail to deliver business value. The essence of AI success is rooted in the seamless integration of AI technologies that harmonize with the fundamental values of an organization. AI solutions should not seek to redefine a business’s distinctive value proposition but rather aim to enhance it.

For businesses aiming to achieve AI-driven, enhanced operational competencies and competitive advantages, partnering with an IT and innovation services provider that offers an end-to-end AI integration service is essential. Ideally, this service should encompass an AI opportunity assessment, solution validation, production, and model management. This strategic approach ensures that AI becomes a cornerstone of business growth and success. In a world where AI is reshaping industries and business landscapes, making the right AI choices becomes paramount for achieving differentiation and competitive advantages.

Armin Haller, serving as the Director for the APAC Centre of Excellence in Data & AI at Crayon, oversees a team of experts dedicated to generating value for clients through the implementation of cutting-edge AI models to address intricate challenges.

Alongside his role at Crayon, Armin holds the position of Associate Professor (on leave) at the School of Computing at the Australian National University. In this capacity, he continues to supervise multiple PhD students, covering diverse topics such as knowledge graph engineering and machine learning. Armin has published over 80 articles in scientific journals and conferences. He also chaired a W3C working group on an ontology standard for the Internet of Things (Semantic Sensor Network Ontology).

TNGlobal INSIDER publishes contributions relevant to entrepreneurship and innovation. You may submit your own original or published contributions subject to editorial discretion.

AI in clinical trials: A game-changer with automation and machine learning