AI might make workers faster, but not necessarily more productive: ‘They do it faster, then go for coffee breaks"
Many companies, spurred by the post-ChatGPT surge in interest, are working to integrate AI into their daily operations.
Generative AI is being hailed as a breakthrough that could bring automation to service and knowledge-based roles—from call centers to consulting firms. But whether using AI to speed up tasks like drafting emails or creating PowerPoint presentations actually boosts productivity remains an open question. Speaking at the Fortune Brainstorm AI conference in Singapore last week, Ramine Tinati, who leads Accenture’s APAC Center for Advanced AI, expressed skepticism.
“If you give employees a tool to do things faster, they’ll do it faster. But does that mean they’re more productive? Probably not, because they just finish early and take a coffee break,” Tinati said.
True gains, he argued, come from rethinking how work is structured: “If you reinvent the work, then those coffee breaks stop being meaningful—because the nature of the work has changed.” He also noted that some companies in Asia have been slower to adopt AI because they focus on improving efficiency without reimagining workflows.
Businesses have been applying AI to boost productivity well before ChatGPT launched in late 2022. For instance, May Yap, chief information officer at manufacturing solutions provider Jabil, shared how her company had implemented AI and automation to support their inspection team—nicknamed the “Golden Eye”—responsible for identifying scratches and defects on phones.
These employees inspect devices for up to eight hours a day, a routine that Yap said inevitably leads to errors. AI was introduced to support the process and help catch mistakes, enhancing the accuracy of inspections.
Similarly, Chee Wee Ang, chief AI officer at Singapore’s Home Team Science and Technology Agency—which develops national security technologies—explained how AI has significantly improved operational processes. “We’ve seen information extraction improve by about 200%. That’s a strong return on investment,” Ang said.
He added that beyond productivity gains, AI now enables the Home Team—which includes agencies such as the police, emergency services, and immigration—to respond to new threats and emergencies that were previously difficult to manage.
Reskilling and Workforce Shifts
Despite AI’s promise, some roles will inevitably become obsolete, raising concerns among employees about being replaced. There’s growing anxiety that workers are effectively training the systems that may one day automate their roles.
At the conference, several panelists agreed that reskilling is the clearest path forward for workers whose jobs are changing.
“Transformation is scary,” said Yap. “When people hear the word ‘transformation,’ they get nervous.” She emphasized that Jabil’s intention is to support, not replace, its human workforce. Skills such as leadership and adaptability, she noted, remain essential and cannot be replicated by AI.
Ang echoed this, noting the challenge of hiring AI-savvy talent in Singapore. His agency has often brought in people with related skills and trained them on the job. He also mentioned hardware constraints, saying that the Home Team relies on on-premise processors due to the sensitive nature of its work—and faces limitations with access to high-performance GPUs.
Tinati remains hopeful about AI’s long-term impact. Rather than replacing people, he sees AI as a tool to elevate them. “Their skills are being redirected toward more valuable tasks—whether it’s supervisory roles or learning new capabilities to support higher-level functions in the development cycle,” he said.