Last month (on November 21, 2024), we hosted a webinar on Artificial Intelligence (AI) for business growth with the theme: “AI for Business Growth - Driving Digital Transformation through AI Integration”. Moderated by Niyi Yusuf, Managing Partner at Verraki, the panel session featured industry experts including Niyi Tayo, Senior Partner, Technology at Verraki; Antony John, Vice President at Omnex Systems; and Idris Musa GM, Information Technology and Asset Integrity (IT&AI) at Oando Energy Resources. The panelists shared insights from real-world AI implementations, highlighting best practices for integrating AI to drive digital transformation in today's dynamic business environment. Here are our top twelve takeaways from the webinar:

AI’s Practicality and Potential

AI is no longer confined to research or specialised settings; it has found practical applications across various sectors, including oil & gas, where it’s advanced analytics optimises exploration and drilling by identifying resource-rich locations with greater accuracy, while AI-driven monitoring ensures compliance and safety. This widespread integration shows that AI can significantly enhance everyday processes, benefiting both businesses and governments. According to UNDP, AI is projected to add about $15 trillion to the world’s GDP by 2030, and half to come from increased productivity. Similarly, Gartner projected that 36% organisations worldwide will be in the experimentation stage with AI by 2027 and will start to adopt use cases with high business value.

Aligning AI with Business Objectives

Successful AI initiatives must align with the broader business objectives and company mission. Leadership buy-in is especially important in markets where younger, digital-native employees work alongside less tech-savvy leaders. AI transformation should be supported by a cohesive strategy that ensures AI technologies contribute to the company's overall goals and customer outcomes. AI initiatives should begin with a clear mission and vision that supports the organisation’s customer-focused outcomes. Engaging leadership to ensure AI initiatives align with strategic goals is essential. A bottom-up approach that involves gathering ideas from teams can foster creativity, while cascading AI objectives throughout the organisation ensures consistent execution across all levels. “Technology investments must align with business objectives to achieve meaningful transformation.” – Olaniyi Yusuf, Verraki

Organisational Readiness and Cultural Shift

Organisations must progress through various stages of AI maturity: from "AI Noise" to becoming "AI Ready," "AI Provisioned," and ultimately "AI Advanced." Cultural readiness, driven by leadership, is crucial for successful AI adoption. Organisations with strong readiness can achieve up to 32% higher ROI on AI investments, showcasing the importance of leadership engagement and fostering awareness across all levels.

Data as the Foundation of AI

Data is the backbone of AI, and quality data is critical for AI models to generate accurate insights. Organisations must focus on robust data strategies, including data acquisition, validation, and processing. Implementing human oversight helps ensure data accuracy, reduce biases, and improve the effectiveness of AI models, mitigating the risk of "AI hallucinations”. To maximise the effectiveness of AI, organisations must prioritise digitising and structuring their data. Well-organised, structured data allows for seamless automation and AI integration, unlocking the full potential of AI technologies in streamlining operations and driving innovation. "Data is the oil of the digital age. AI is allowing us to process and utilise this resource faster than ever before." – Idris Musa, Oando

Responsible AI Implementation

Responsible AI practices focus on minimising biases, ensuring transparency, and building trust with users. This includes using diverse and inclusive datasets and making AI systems reliable and accessible. Regional considerations are also important, especially in underserved areas where AI systems must be effective across diverse cultural contexts. Robust and diverse African data is necessary for AI to serve the continent fairly. "African data must be robust and diverse to ensure AI serves the continent fairly and effectively." – Niyi Tayo

Critical Success Factors for AI Deployment

The success of AI deployment hinges on several factors. People and culture are key, requiring active leadership to champion AI adoption and encourage cross-functional collaboration. Value-driven implementation ensures AI investments are aligned with measurable business outcomes, while strategic partnerships can foster shared learning. Finally, high-quality data is essential for accurate AI results and to prevent issues like data bias. "To achieve higher ROI on AI investments, cultural readiness, data quality, and leadership alignment are non-negotiable starting points." – Antony John, VP at Omnex Systems

Building an Effective AI Strategy

A well-defined AI strategy begins with a clear vision and roadmap that articulates the organisational goals for AI implementation. A thorough maturity assessment helps identify readiness and any gaps that need addressing before deployment. Rapid scaling through pilot programs ensures AI solutions are scalable and deliver value quickly, allowing organisations to refine strategies as they grow.

Choosing the Right AI Model is Critical

Selecting the appropriate AI model is crucial to meeting specific business needs. Not all AI solutions require complex generative models, so businesses must assess factors like data volume, variability, and cost when choosing AI technology including NLP, machine learning, or generative AI. Making the right choice ensures sustainable, cost-effective AI implementation.

Emerging Solutions and Industry Practices

Human-in-the-loop models, which involve experts validating AI outputs, help ensure reliability and accuracy. Advanced tools like vectorised databases and fine-tuning techniques reduce bias and enhance model performance. The use of LLM Jury Systems, which employs multiple language models to validate AI outputs, further ensures higher accuracy and reduces the risk of errors or biases in decision-making.

Key Challenges in AI Implementation

Several challenges can hinder AI adoption, including resistance to change, data availability, and transparency. Addressing biases in AI models, particularly in generative AI, is essential for building trust. The high cost of domain-specific training and the need for explainable AI models also pose significant hurdles. Overcoming these challenges requires careful planning and rigorous testing to ensure AI solutions are effective and equitable.

The Cost of Inaction

The failure to adopt AI can result in significant losses for businesses, from losing clients, poor customer experience to declining workforce productivity and diminishing market relevance. Understanding the cost of inaction highlights the importance of modernising operations through AI, as businesses that lag risk falling out of step with industry advancements and missing growth opportunities.

AI Readiness Assessment

Organisations must develop their vision for the use of AI and then conduct a readiness assessment to understand the gap between their AI ambition and their current AI capabilities and be able to define a pathway for bridging the gap. Verraki’s AI Readiness Assessment Toolkit helps organisations evaluate their AI readiness across twelve key dimensions, providing actionable insights into how the organisation operates, organises and behaves. The toolkit focuses on critical factors such as operations, technology infrastructure, work organisation, governance, policies, performance management and corporate culture, helping businesses optimise their AI adoption strategy and improve overall success.

To watch a replay of the webinar, please click this LINK.