Skills, workflows and governance: What businesses need to make AI work

Transitioning to artificial intelligence (AI) in the workplace is a complex process that involves rethinking jobs, employee support and technology guardrails, according to Paul Carvouni, Salesforce Asean senior vice-president and general manager. Singapore has been proactive in embracing new technologies, with initiatives such as the National AI Strategy 2.0, but the integration of AI into the workplace may be slower than expected.

A Salesforce survey found that 97% of workers anticipate working with AI agents, suggesting uncertainty about the impact on individual jobs and the skills needed to collaborate with AI.

To bridge this gap, businesses must invest in workforce skills, redesign workflows, and establish trust and governance. Basic AI literacy is a start, but employees need to develop skills that combine their professional expertise with an understanding of how AI can be applied to their jobs. Singapore's National AI Impact Programme is working to help companies adopt AI while equipping workers with the necessary skills.

Businesses can complement these efforts by moving beyond one-off training sessions and demonstrating how AI can be incorporated into daily tasks. This may also involve revising job scopes and team structures as people work more closely with AI agents.

Redesigning workflows is crucial for realizing the full potential of AI. By identifying routine tasks that can be handled by AI and determining where human judgment, expertise, and interaction remain essential, businesses can achieve significant improvements. For instance, customer service representatives in insurance could use AI to help draft responses, while an AI agent could manage routine queries and escalations, leading to substantial time savings.

In the Philippines, health maintenance organization Maxicare has implemented this approach for its Letter of Authorisation process, reducing service time by 97.5%.

Trust and governance are also essential considerations. Customers must be assured that their data is protected, and employees need clarity on how AI systems are being used. Regulators expect businesses to be accountable for their AI systems and processes, particularly with large language models (LLMs) whose responses are probabilistic rather than rule-based.

As AI becomes more widely used across organizations, governance must scale accordingly, allowing businesses to set controls, monitor usage, and manage risks, security, and compliance.

Despite some workers potentially turning to AI tools for personal use, businesses must go beyond treating AI as another technology cycle to wait out. The nature of work is undergoing a generational shift, with AI having the potential to take responsibility for certain tasks and processes, fundamentally changing the role people play within them.

For Singapore, a small domestic labor force and an aging population make it crucial for workers to become more productive alongside AI. The question is not merely whether businesses should adopt AI but how they can do so in a way that allows their workforce to adapt alongside it.