Unlocking Artificial Intelligence's potential

I began exploring the use of AI tools in our research and analysis back in 2022, a year that saw the rapid rise of generative AI (GenAI) and the popularisation of Large Language Models. While the potential of these tools was undeniable, their tendency to produce hallucinations (fabricated or nonsensical outputs) raised significant concerns about reliability and accuracy.

This, combined with issues surrounding data sovereignty and privacy, prompted BERL to take a cautious and strategic approach. We restricted the use of AI tools to ensure compliance with data governance standards and to safeguard sensitive information. Additionally, many of the tailored AI solutions available were prohibitively expensive, designed primarily for larger organisations, making them unsuitable for our needs.

As part of my Master of Technological Futures programme with academyEX over 2023-24, I wrote a paper using BERL as a case study to address these challenges. I explored critical areas such as strategic alignment, ethics, technology requirements, use cases, privacy considerations, data governance, staff training, and success measurement.

The process was iterative and collaborative, with our team serving as early adopters and testers. I am deeply grateful for their willingness to experiment and adapt. This roadmap is now being offered freely as part of BERL's pro bono work to benefit resource-constrained SMEs and NGOs, enabling them to navigate the complexities of AI adoption effectively while maintaining ethical and operational integrity.

This roadmap provides a practical guide for SMEs and NGOs in Aotearoa New Zealand to successfully integrate AI into their operations. It outlines a step-by-step approach to strategic alignment, ethical considerations, implementation, and continuous improvement. The roadmap provides a structured approach to leveraging AI for enhanced efficiency, innovation, and competitive advantage, and ensuring any AI adoption is well-defined, strategically aligned, and provides a clear path to achieving long-term business goals.

Here's a high-level summary of the step-by-step guide to the roadmap:

1. Articulate strategic alignment: Define clear objectives for AI adoption and align them with your business strategy and long-term goals
2. Establish leadership mechanisms: Set up leadership structures to oversee AI initiatives, define responsibilities, and ensure accountability
3. Conduct a Privacy Impact Assessment: Identify and mitigate privacy risks, ensure compliance with privacy laws, and build trust with stakeholders
4. Identify use cases: Find specific areas where AI can add value, focusing on operational efficiencies, quality assurance, and innovation
5. Understand technological needs: Evaluate and improve the technology needed for AI initiatives, including data management and computational resources
6. Understand capability needs: Identify and acquire the necessary talent and skills for AI, and implement training and upskilling programmes
7. Ensure ethical and responsible AI practices: Incorporate ethical considerations, address biases, and ensure compliance with regulations and standards
8. Measure success and progress: Develop metrics to track the success of AI initiatives, evaluate effectiveness, and make evidence-based decisions
9. Implement the digital strategy: Execute the strategy, address infrastructure requirements, overcome data challenges, and deploy AI solutions incrementally while promoting learning and improvement.

Click here to read the roadmap in more detail.