Over the past 18 months, the introduction of powerful artificial intelligence (AI) tools within the public mainstream has pushed businesses to reassess their processes in an effort to streamline their operations and accentuate efficiencies through the adoption of the technology.

However, in the rush to embrace the potential AI could afford, many are setting themselves up for disappointment and unnecessary expense, says Lincoln Grima, Chief Commercial Officer at ICT Solutions. In his words, “AI must be treated as a strategic and governed capability.” Without that framing, even the best tools become costly experiments, he insists.

This caution comes as the excitement around AI has convinced leadership teams, the C-suite and boards of directors to, at times, make hasty decisions that sideline preparation in favour of quicker implementation. Indeed, Mr Grima has seen it repeatedly: “Leadership teams and management [are] deciding to jump on the bandwagon without firstly defining a clear AI strategy, and without being able to educate their employees on the best practices around using AI ethically and responsibly.” The consequences, he continues, are predictable – patchy execution, data exposure, and missed value.

Crucially, sustainable adoption requires a clear plan of enablement and a governance structure that can accommodate flexibility. “If you simply give a licence to your staff without going through the correct adoption approach, your team might decide not to use it in those situations where its deployment could be truly beneficial,” Mr Grima says, warning that buying licences and expecting success is one of the fastest ways to burn money and become sceptical of the value.

In other words, AI implementation is not simply a case of enabling tools such as Microsoft Copilot and Google Gemini. While this may seem to be an obvious first step – with companies expecting an immediate return on investment from the new tools – employees have typically not been adequately trained to understand how AI can add value to their workflows. “Most enterprises have gone through – and are still going through – a period of adaptation and yet they often do not have the internal expertise to adopt AI in the right way,” Mr Grima attests.

ICT Solutions' Chief Commercial Officer, Lincoln Grima

Fundamentally, sustainable adoption starts with intent, he underscores. “There are industry-standard AI adoption frameworks that have been purposely designed to embed AI within a company’s strategy and culture,” stresses Mr Grima. To put it differently, AI needs to be tied to concrete business outcomes, owned by leaders, and reinforced through process, rather than being treated as a side project or as an optional tool.

This strategic spine encompasses the entire lifecycle of the adoption process, from use case selection and risk assessment to operating model, change management, and measurement, Mr Grima insists. This implies the necessity to foreground a holistic business plan in order to avoid isolated wins that, on their own, might lead to system wide friction. “Resist the urge to simply provide AI to the team without embarking on a sustainable adoption journey,” Mr Grima insists.

One of the risks of a lack of proper AI strategy concerns security. “It is important to employ the best-in-class practices for cybersecurity within an enterprise’s data structure and governance to make sure the business is not mistakenly leaking sensitive corporate data,” Mr Grima stresses. The most avoidable error, he insists, is copy pasting sensitive information into public tools. “Inputting data in free models means that this data is being used to train those models,” he cautions.

Moreover, skill-building – for instance with regard to prompting – is a core aspect of a sustainable AI adoption strategy. “Prompting has become a valuable tool to get the required value out of generative AI,” Grima notes, adding that employees also need to be trained in problem decomposition, verification practices, and workflow integration. This is an executive responsibility, with the adoption of AI not constituting just an IT rollout. “Whether C suite executives or management are providing AI training to their employees or simply giving them the licences for tools, it is important to understand how to adopt AI in the right way, and how to use Gen AI as part of a business strategy.”

The speed of change is also a consideration, the CCO attests, explaining how Agentic AI will soon become the new technological frontier, and urging businesses to think ahead. “The crucial difference here is that Gen AI creates content in response to users’ prompts, while Agentic AI actions tasks and makes decisions without needing a direct prompt from the users,” Mr Grima explains. Some agents run autonomously; others keep humans in the loop for oversight and accuracy. However, few companies are there yet. “There are very few companies currently benefiting from Agentic AI. They are still adopting the first layer of Gen AI,” Mr Grima observes.

Yet, such technology can benefit a range of industries, the CCO explains. For example, in financial services, agents can orchestrate fraud signals across systems, run automated AML/KYC checks against policy, and escalate exceptions with documented rationale. In logistics, agents can continuously re plan routes as constraints change, generate and validate contract terms against playbooks, and flag anomalies in shipment data – without waiting for a prompt. In customer operations, agents can triage and resolve requests end to end across channels, pulling knowledge, filling forms, triggering workflows, and invoking humans only when confidence drops below a threshold.

These are not distant hypotheticals, Mr Grima asserts, but the logical next step once guardrails, data access, and process integration are in place. “Agentic AI is the point where the technology becomes very impressive, helping companies become scalable through the introduction of independent AI agents who will become copilots,” he explains.

He offers a simple metaphor: “The driver has the responsibility to keep the car on the road and to get to the finish line as fast as possible, but the copilot is there to guide and assist the driver through that process.” To rephrase, while the human driver remains accountable, the agent bolsters capability.

Concluding, Mr Grima urges businesses to begin their AI journey sooner rather than later. “Those businesses – of any size – that will not start their journey towards a sustainable AI adoption will be late to the party and will find it hard to keep up with their competition,” he cautions.

The imperative is clear: “Now is the time to adopt AI seriously and place it as a key strategic pillar within any organisation – whatever the industry.”

ICT Solutions is a leading managed services provider focused on AI adoption & cybersecurity to support Maltese industry. They can be contacted at [email protected].