A Future-Proof AI Strategy Requires Build-for-Purpose Enterprise AI Solutions

In today’s dynamic business landscape, integrating artificial intelligence (AI) into core operations is not only advantageous but essential for long-term success. At the recent AI Expo Africa 2024, held in Johannesburg, Len Pienaar, Managing Director at AdvanceGuidance, shared insights on the real reasons so many corporate AI projects fail. 

Pienaar explored aspects of AI Integration in his session titled ‘A Case Study in Innovation & Strategy: A Built–for–Purpose Enterprise AI Solution.

Reflecting on a proof-of-concept project deploying a Generative AI leveraging a Large Language Model (LLM) for a global consulting firm, Len discussed the real challenges of deploying build-for-purpose enterprise AI solutions to enhance operations, to ensure compliance, and provide a competitive edge to future-ready businesses.

AI Project Failure Rates High

Today, according to Pienaar, over 80% of corporate AI projects fail, leading to accusations of AI merely being hype.

“Generic AI services often fail to address the unique challenges faced by enterprises”, Says Pienaar, “The true power of AI for business lies not in generic solutions but within a strategy that deploys build-for-purpose enterprise AI tailored to the unique needs and challenges of individual businesses” he commented. “My experience across diverse sectors—banking, telecommunications, and media—has shown me first-hand the transformative potential of these technologies.”

Growing Demand for AI in Enterprises

The global enterprise AI market is experiencing unprecedented growth, with projections estimating its value will reach $407 billion by 2027². This surge reflects a broader trend across various industries, including finance, telecommunications, healthcare, and manufacturing.

Recent research commissioned by IBM, reveals that 42% of enterprise-scale organisations actively use AI, and an additional 59% plan to increase their investments in this technology.

A striking 83% of companies now place AI at the forefront of their strategic planning.

Early adopters are reaping substantial benefits from their AI initiatives, demonstrating that overcoming the initial deployment barriers can lead to significant returns.

Build-for-Purpose AI Solutions Essential

Organisations must focus on build-for-purpose enterprise AI solutions designed to automate complex, repetitive tasks while ensuring high-quality outputs according to Pienaar.

These systems continuously learn from data, refine their operations, and enhance performance with a high degree of autonomy—often capable of functioning without direct human oversight. 

Says Pienaar, “Build-for-Purpose enterprise AI solutions are capable of automating most tasks performed by highly skilled, trained, experienced professionals today, allowing these scarce resources to focus on more strategic activities”.  “Although it is still very early days, I am specifically excited about the potential of Generative AI using Large Language models in this domain.” he expands. 

While the potential for AI is vast, implementing these technologies comes with significant technical challenges—especially in highly regulated industries where compliance, security, and data privacy are vital.

AI Project Risk Factors

In addition, organisations need to navigate hurdles related to functionality, scalability, and quality. However, Pienaar remarks, “Given access to capable teams of skilled, AI-experienced people, technical challenges are not why these projects fail. Addressing the softer aspects of the projects, relating to people, often makes the difference.”

The high failure rate of corporate AI projects today is primarily due to several non-technical factors:

• A misalignment of business goals,
• poor quality or legacy data,
• A focus on the latest technology rather than solving real business problems,
• And unrealistic and uninformed expectations about AI’s capabilities.

Concludes Pienaar, “Embracing a skills-first, human-centric approach, which focuses on investing in access to people with experience in AI, fostering collaboration between AI systems and human workers, and prioritizing the human aspects of technological integration, will enable organisations to transition from mere buzzwords around AI to tangible breakthroughs that drive success.”