Skip to main content
Contact us
Crossing the AI chasm – how CTOs can drive adoption
Share on socials

Crossing the AI chasm – how CTOs can drive meaningful adoption in legacy landscapes

Matt Saunders
Matt Saunders
Published on August 4, 2025
9 min read
AI brain with yellow lightbulb above it
Matt Saunders
Matt Saunders
Published on August 4, 2025
9 min read
Jump to section
Introduction
Integrating AI: What are the challenges?
Governance and risk managemen
The talent gap persists
Taking a strategic approach
Make AI part of everyone's day-to-day
The path forward
Frequently asked questions about adopting AI

A practical guide for integrating AI into your workflows and legacy systems, helping to balance transformative AI adoption with operational stability as part of digital transformation.

The artificial intelligence revolution has reached a point where it's clear that most enterprises should engage in it or risk being out-innovated by competitors who do. However, according to recent McKinsey research, organisations using AI report measurable business value, yet many struggle with implementation challenges, particularly when integrating new AI capabilities with legacy systems.
For CTOs, the revolution isn't just technical; it's about balancing introducing potentially game-changing but disruptive AI strategies into a business while keeping it operationally stable.

Integrating AI: What are the challenges?

There's clearly a big tension between the pace of AI adoption and the collective need to understand how and when a return on investment will emerge. Not to mention how this disruptive technology fits around the other teams and technologies in the organisation. This is particularly important for convincing decision-makers to invest in cutting-edge and potentially expensive technology to sit alongside legacy tools. With the pace of change in the AI world, almost every business now has this challenge.
The real power of generative AI lies in giving it enough contextual information to provide authoritative answers based on the organisation's cumulative knowledge. For most organisations, the information most valuable to an AI tool is spread across many technologies and data stores. Teams winning at this are investigating how to connect legacy systems that weren't designed for contemporary AI consumption. They're also using emerging RAG (Retrieval-Augmented Generation) technology to connect these systems.

Governance and risk management are critical success factors

Ensuring that AI has the correct level of access to what could be personal or privileged data is a complex problem, and governments are already legislating to ensure that businesses take it seriously. The European Union's AI Act, which came into effect in August 2024, has established new compliance requirements that organisations must navigate carefully.
Businesses need a detailed AI governance framework outlining the ethical standards people should follow, how data is used, and how AI will be held accountable. Issues such as algorithmic bias, data privacy, and model explainability are of note here, and ensuring these are covered is key.

The talent gap persists

There's a shortage of experienced AI professionals. Analysis from the UK Government's Office for AI highlights a persistent skills gap in machine learning and data science roles. The pace of change has accelerated dramatically with the generative AI revolution.
While some employees will readily embrace new tools that increase productivity and automate manual tasks, others will be sceptical and mistrust AI. They might even worry they're being replaced. These concerns could impact adoption and cause internal pushback against AI tools. Looking further, our recent report Digital Etiquette: Unlocking the AI Gates, found that the majority of AI training is offered to high earners, and organisations must actively try to redress this balance to avoid AI expertise resting in the hands of the few.
Successful organisations are adopting hybrid approaches: hiring core AI specialists and developing existing technical staff through structured learning programmes. This approach maintains institutional knowledge while building new capabilities.

Successful organisations are taking a strategic approach

Organisations winning at AI are adopting a measured, risk-managed approach to integration within existing organisational structures. Here are some successful strategies we've noticed:
  • Aim to balance competitive urgency with strategic planning by building compelling business cases that clearly align AI investments and long-term organisational objectives.
  • Assess your data readiness and quality, secure appropriate talent and expertise, establish comprehensive governance protocols for ethical and regulatory compliance, and prepare your workforce for technological change as foundational requirements.
  • Follow a structured implementation progression by starting with small-scale pilot projects in low-risk, high-impact areas to test value propositions before scaling organisation-wide.
  • Define measurable KPIs aligned with business outcomes, engage stakeholders through education and communication initiatives, and develop robust governance frameworks that address data strategy and ethical standards.
  • Learn continuously and adapt by utilising insights from pilot programmes to inform broader organisational rollouts. Focus on workforce development by implementing role-specific training, developing AI literacy programmes, and cultivating internal champions who will drive adoption and provide ongoing support throughout your digital transformation process.

Make AI part of everyone's day-to-day

As with any workplace initiative, the biggest consideration is your people. That's why helping yours to stay ahead of LLM-driven changes is so important to the success of your AI efforts. Don't be complacent and hope that everyone will catch up on their own. Instead, make sure you:
  • Offer role-specific training, tailoring learning opportunities to allow people from different teams to learn how AI and LLMs can enhance their work. Make sure this doesn't just go to a select few high-earners if you want the whole organisation to benefit.
  • Encourage AI literacy by running organisation-wide workshops on AI fundamentals, including the ethical and governance considerations, to build strong foundational knowledge.
  • Create AI champions. These early adopters can act as in-house experts and change agents, helping with adoption and providing support. AI centres of excellence that combine technical expertise with business understanding provide crucial guidance on tool selection, implementation support, and ongoing maintenance.
  • Normalise the use of AI. Where there's an option to integrate AI functionality into an existing platform easily, take it. Get people familiar with using the functionality with easy wins. Business users need an understanding of AI capabilities and limitations to identify appropriate use cases.
Keep them interested: set learning goals, provide course access, and reward progress to incentivise your people to keep learning more. This could also include AI certifications.

The path forward

Generative AI adoption has increased massively over the past two years, offering valuable implementation lessons. AI integration challenges are, however, becoming more standardised, with established patterns and solutions emerging across industries. CTOs can leverage these emerging best practices while adapting them to their own organisational contexts.
Successful AI integration in legacy environments needs a balanced approach that simultaneously addresses technical, organisational, and strategic considerations. CTOs must resist the pressure to implement AI solutions hastily and instead focus on building sustainable capabilities that deliver measurable business value.
Organisations that will succeed in AI adoption treat it as a comprehensive digital transformation initiative rather than a technology implementation project. This approach requires a commitment to long-term capability-building rather than quick wins.
By focusing on building foundations with a strategic implementation and continuously developing organisational capabilities bolstered by AI, CTOs can bring in long-term change if they avoid the hype and look to solutions that help deliver incremental but gradual transformational change, leveraging AI.
If you're interested in finding out more about how you can best build this balance into your AI strategy, get in touch with our digital transformation experts today.
Adaptavist can help you find the right tools to align with your business needs, implement them to optimise their value, and support your teams with training and working practices so they can start using AI with confidence.

Frequently asked questions about adopting AI

CTOs should start with small-scale pilot projects in low-risk, high-impact areas to test AI capabilities before scaling. Use RAG (Retrieval-Augmented Generation) technology to connect legacy systems with modern AI tools, ensuring proper data flow and contextual information access. Focus on building robust governance frameworks and establishing clear KPIs aligned with business outcomes to measure success and guide broader implementation.
Organisations can address the AI talent gap through a hybrid approach: hiring core AI specialists while developing existing technical staff through structured learning programs. Successful strategies include offering role-specific AI training tailored to different teams, implementing organisation-wide AI literacy workshops, creating AI champions as internal experts, and establishing AI centres of excellence. This approach maintains institutional knowledge while building new AI capabilities across the organisation.
Organisations need comprehensive AI governance protocols that address data privacy, algorithmic bias, and regulatory compliance, including the EU AI Act. Essential elements include ethical AI standards, clear data usage guidelines, model explainability requirements, and accountability measures. Establish risk assessment procedures, regular auditing processes, and transparency protocols, especially when handling personal or privileged organisational data.
Follow a structured implementation progression: begin with pilot projects that demonstrate clear business value, then use insights to inform broader rollouts. Develop AI centres of excellence combining technical expertise with business understanding. Focus on continuous learning and adaptation, establishing feedback loops between pilot programs and organisational strategy. Ensure proper change management by engaging stakeholders through education and building sustainable AI capabilities rather than pursuing quick wins.
Combat AI resistance through role-specific training programs and organisation-wide AI literacy initiatives. Create AI champions as internal advocates and change agents to support adoption. Address job displacement concerns directly through transparent communication about AI's role in enhancing rather than replacing human work. Ensure equitable access to AI training across all organisational levels, not just high earners.
Written by
Matt Saunders
Matt Saunders
DevOps Lead
From a background as a Linux sysadmin, Matt is an authority in all things DevOps. At Adaptavist and beyond, he champions DevOps ways of working, helping teams maximise people, process and technology to deliver software efficiently and safely.
Digital transformation
;