BillionPhotos.com - Fotolia

Happy New Year: Three steps to maximise AI’s ROI in 2025

On the Eighth Day of AI, we explore the key considerations and strategic frameworks essential for extracting maximum value from AI projects.

According to Gartner research 85% of all AI projects fail, and if you’re a business leader looking to drive value from technology transformation and leverage AI, this wouldn’t fill you with confidence.

But AI does have the potential to revolutionise industries, and its adoption isn’t set to slow down. In fact, global AI adoption by organisations is set to expand at a CAGR of 36.6% between 2024 and 2030. When effectively adopted and utilised, the benefits of AI are immense.

The reason many AI projects fall short is because they have unclear objectives, inadequate measurement frameworks, and an unstructured approach to implementation. As we enter a new year, UK businesses must strategically harness AI’s full potential to remain competitive. To maximise the benefits of AI in 2025, business leaders should follow three critical steps: assessing AI readiness, building trust, and establishing robust governance.

Step one: Laying the groundwork for AI readiness

  • Preparing for AI integration

 The journey towards successful AI integration begins with readiness. Preparing for AI involves more than just adopting new technologies; it requires a comprehensive strategy that includes defining AI policies, scalable technology tools, and robust governance frameworks. Establishing these foundations ensures that AI initiatives are not only effective but also sustainable in the long term.

  • Assessing AI readiness

AI readiness encompasses both technical and organisational preparedness. It involves assessing the current state of data infrastructure, talent, and organisational culture. A key first step is conducting a thorough AI readiness assessment to identify gaps and opportunities. This assessment should cover:

  • Data quality and availability
  • Existing technological capabilities
  • Workforce readiness to embrace AI-driven change

Setting clear objectives and KPIs against which to measure progress is essential for driving focus and accountability. For instance, Microsoft has reported that 70% of Copilot users said it makes them more productive. So, if a company had set out productivity as a key metric for success, their AI project would definitely have been effective.

Once you’ve created a solid foundation, it’s time to focus on successful implementation.

Step two: Implement an AI strategy built on trust

  • Building trust in AI

Trust is paramount in AI adoption. Without trust, the potential benefits of AI cannot be fully realised. According to a KPMG study, 73% of people express concerns about the risks of AI, while 75% are more willing to trust AI when assurance mechanisms for ethical and responsible use are in place.

  • Emphasising transparency, accountability, and explainability

 Building trust in AI requires a commitment to transparency, accountability, and explainability. Transparency involves making the decision-making processes of AI systems visible and understandable to stakeholders. This can be achieved through clear documentation and communication about how AI models are developed and deployed.

Accountability ensures that AI systems are responsible for their actions and decisions. There must be a clear chain of responsibility for any adverse consequences, ensuring that developers and operators are accountable. Explainability goes a step further by making the AI's decision-making process understandable to end-users. This not only builds trust but also enables stakeholders to effectively use and manage AI systems.

Once you’ve implemented AI into your organisation, you need to put the guardrails in place to ensure responsible and effective use of AI on an ongoing basis.

Step three: Establish robust governance frameworks

  • Addressing ethical considerations

Whether business leaders are adopting or creating their own AI solutions, they need to make sure they are prioritising ethical considerations. This means sense checking both the data that is going into AI and AI outputs for potential biases and ensuring fairness. Ethical AI adoption also requires protecting data privacy and security, safeguarding personal information against cyber threats and data breaches.

  • Establishing governance frameworks

Effective AI governance is crucial for balancing innovation with ethical and societal considerations. Governance frameworks should include regulatory compliance, ethical standards, and accountability mechanisms.

To effectively govern AI initiatives on an ongoing basis, businesses should establish AI governance boards. These boards are responsible for overseeing AI strategies, policies, and compliance. They play a crucial role in ensuring that AI projects align with organisational goals and ethical standards.

  • Defining responsibilities

The AI governance board should include stakeholders from various departments, including IT, legal, compliance, and business units. Their responsibilities include defining AI policies, monitoring AI deployments, and ensuring adherence to regulatory and ethical standards. Additionally, they should facilitate continuous education and training to keep the workforce informed about AI advancements and ethical considerations.

The 12 Days of AI

A happy New Year

By strategically navigating the integration of AI in 2025, UK businesses can unlock unprecedented opportunities for innovation and growth. But this requires a thoughtful approach that prioritises readiness, trust, governance, and responsible development. Through these efforts, businesses can ensure that their AI initiatives are not only effective but also aligned with broader societal and ethical values. But if business leaders follow the three steps outlined, they can get the New Year fireworks and sparklers ready and pave the way for a successful and sustainable AI-driven future.

Robert Cottrill is technology director at ANS, a digital transformation provider and Microsoft’s UK Services Partner of the Year 2024. Headquartered in Manchester, it offers public and private cloud, security, business applications, low code, and data services to thousands of customers, from enterprise to SMB and public sector organisations.

Read more on Artificial intelligence, automation and robotics