Driving AI Adoption: unpacking key recommendations

Drivers behind low levels of AI adoption

In our work with clients across industries, one pattern surfaces repeatedly: organizations are often racing to experiment with AI – from Traditional AI to GenAI and Agentic AI – with siloed projects and without a coordinated adoption plan. The result? Low uptake, unrealized benefits, and disappointing returns on investment.

The common denominator is clear: Change management is an after-thought. Too many teams focus on deploying the technology as quickly as possible, without preparing the people, processes, and culture that determine whether AI sticks.

For risk-averse organizations, the urgency to “do something with AI” can override the need for a thoughtful adoption strategy; one that addresses stakeholder engagement, cross-functional alignment, and operational readiness.

To get AI adoption right, CIOs must look far beyond the tech stack. It’s about embedding change management from day one, ensuring every impacted process and stakeholder is part of the journey. This is of course true for all technologies (not just AI) but with AI’s emerging unit / token economics, this becomes even more important.

Ensuring lasting AI adoption

To ensure effective AI adoption, we recommend designing and implementing an appropriate AI adoption strategy, with guidance across the following three (3) topics:

1. AI Communication Strategy
2. AI Literacy
3. Culture

1. AI Communication Strategy

The information required by each team involved across the AI Product Lifecycle, and at different stages of the process, differs greatly.

Naturally, leadership teams seek information around expected business value and realized benefits; while risk & compliance teams are interested in identifying potential risks and defined controls.

Meanwhile, end-users want to clearly see examples of how deployed AI use cases can be used in their roles to improve efficiencies or simplify processes.

Therefore, we recommend that communications are designed around AI initiatives that are tailored to the needs of each target audience group. Furthermore, communication plans should specify approach and communication milestones throughout each stage of the AI product lifecycle – from strategy setting to AI use cases experimentation, deployment at scale and operationalization.

Read how the development of storytelling assets helped a Financial Services client capitalize on AI.

2. AI Literacy

The skills, knowledge and understanding necessary to effectively (and safely) deploy and interact with AI systems is rapidly evolving.

This is driving organizations to assess their current AI literacy levels and evaluate their current practices. The aim should be towards upskilling and future-proofing the organization’s talent, as well as promoting trust in AI Adoption through a better understanding of AI capabilities and limitations, potential risks and ethical considerations.

Regulatory obligations

AI literacy is no longer a “nice to have”, evolving from good practice to a regulatory obligation (e.g. EU’s AI Act article 4 – AI Literacy). Leaders across various sectors are increasingly stressing it is vital for innovation and strategic competitive advantage. Organizations are taking different approaches with varied levels of urgency.

Tailoring external training

Mature institutions are regularly curating publicly available AI trainings, combined with internally-developed trainings for use cases that are too specific to a role or an organizational need.

Additionally, companies are partnering with universities, educational platforms and training service providers to create comprehensive learning pathways or cover identified skill gaps.

It goes without saying that AI up-skilling campaigns must be engaging and tailored to audience roles and AI competency levels. With micro-learning modules that facilitate consumption within reduced timeslots as well as greater flexibility to adapt/update modules with a fast-evolving topic such as AI.

HR ramifications

Changes will also be required in the way leadership teams define performance evaluation and remuneration; with employees being expected to complete a defined number of AI training modules as well as effectively demonstrating understanding and appropriate usage of AI.

Lastly, AI up-skilling campaigns should encourage experimentation via game-based learning, AI scenario simulations, hands-on sandbox prototyping, AI-tutors assistants and more.

3. Culture

AI Adoption is a human challenge, therefore organizational culture is a primary factor for the success or failure of AI initiatives.

For a robust AI Adoption roadmap, it’s important to assess which organizational and cultural barriers to AI adoption exist within your team and organization (i.e. AI mistrust, skepticism, change aversion, fear of job displacement).

With a better understanding of these barriers, different techniques can be used to overcome them and increase the chances of success.

To improve organizational culture alignment with AI Strategy we have 4 key recommendations

1. Encouraging active experimentation

• Avoid engaging your end-users too late into the journey (e.g. once AI solution is deployed) and expect their buy-in and adoption
• Engage them from ideation & prototyping stage, leveraging their content knowledge to gather feedback throughout the AI product lifecycle
• Enable users to own the solution, this will make adoption all the more permanent

2. Providing access to appropriate AI solutions and tooling

• Increase awareness & familiarity with approved AI tools/apps that are available to use
• Aim towards self-service approach for simple tasks, reducing bottlenecks created by reduced AI SME experts required to respond to growing volume of AI queries & interest
• Minimise the risk of ‘Shadow AI’ (use of AI tools or applications by employees without the required assessment, approvals and oversight by IT, Risk & Compliance teams)

3. Designing secure and user-friendly environments

• AI beginners will be demotivated to use AI solutions if interactions are not simple and user friendly – regardless of how powerful and useful the outputs may be
• User concerns on data privacy, security, accuracy (hallucinations) may deter them on using AI solutions, unless these concerns are acknowledged, mitigated and monitored
• To increase experimentation and adoptions, users need to feel safe using AI tools and applications

4. Fostering a community

• Share ideas, challenges & success stories – promote collaboration across different AI stakeholders (from users to developers, risk, compliance, leadership and more)
• Gather feedback on a regular basis across all stages of AI lifecycle journey – promote early identification of issues and areas for improvements
• Recognise & reward AI experimentation – promote proactivity, adoption and a safe failure culture – rather than rigid ‘failure is not an option’ approach

The bottom line?

Adopting AI isn’t about throwing tech at problems. It’s about building the right environment – where culture, communication, skills, and strategy come together. Get that right, and AI won’t just be another buzzword – it’ll be your competitive advantage.

⁽¹⁾ Source: Wavestone Data & AI Executive Survey (Jan 2024) https://www.wavestone.com/en/news/2024-data-and-ai-leadership-executive-survey-41/ (opens in a new tab)

⁽²⁾ Source: How to Scale GenAI in the Workplace https://sloanreview.mit.edu/article/how-to-scale-genai-in-the-workplace/  (opens ins a new tab)