Jakarta. Indonesian companies risk missing out on artificial intelligence-driven productivity gains by remaining stuck in small-scale pilot projects instead of deploying the technology across their operations, according to a senior Salesforce executive.

Paul Carvouni, Salesforce’s senior vice president and general manager for ASEAN, said businesses needed to move beyond experimentation by investing in workforce training, redesigning operational processes, and strengthening safeguards for AI systems.

Indonesia has rapidly adopted digital technologies, from mobile banking to e-commerce, but turning that uptake into measurable productivity improvements remains a challenge as the country seeks to achieve high-income status by 2045.

The next phase of AI development, involving autonomous systems known as AI agents that can reason and execute tasks, will require companies to rethink how work is organized, Carvouni told the Jakarta Globe.

“Corporate lag” risks leaving employees aware of technological changes but without the skills and tools to use AI effectively, he argued.

Concerns over data privacy, security, accuracy, and reputational damage have encouraged companies to proceed cautiously. However, Carvouni warned that prolonged experimentation could prevent businesses from realizing the technology’s benefits while pushing employees to experiment with AI tools independently, potentially creating additional security risks.

He cited Salesforce survey findings showing that fewer than 3% of Indonesian knowledge workers never expect to work with an AI agent. The finding suggests that resistance to AI may be less widespread than concerns about employees’ awareness of the technology and their preparedness to use it, he added.

Carvouni said businesses needed to develop an “AI bilingual” workforce that combines professional expertise with the ability to work effectively alongside AI systems. Rather than relying on one-off training sessions, companies should provide continuous skills development and adapt job descriptions and organizational structures to accommodate human-AI collaboration.

Companies must also redesign workflows instead of simply adding AI tools to existing processes, he said.

In insurance, for example, AI could move beyond helping human agents draft customer responses to independently handling routine inquiries and taking action, with complex cases transferred to human staff.

Indonesian health insurer Inhealth is using Salesforce’s Agentforce Service platform in its contact center. According to Carvouni, the company reduced processing time for its Letter of Guarantee service by 46% within three months, while chat response times fell by 31% to 24 seconds.

The figures illustrate how AI deployment can improve operational efficiency, although the results come from a company-specific example rather than evidence of economy-wide productivity gains.

Carvouni also stressed the importance of governance as companies expand AI use. Because large language models can produce inconsistent or inaccurate responses, businesses need controls to protect customer data, monitor outputs and ensure accountability, he said.

The challenge is particularly relevant in Indonesia, where companies must serve customers across a vast archipelago and a rapidly evolving digital economy.

For business leaders, Carvouni said, the priority should be to translate AI investment into measurable improvements in service, productivity and growth while preparing employees for changing roles.

Without sustained investment in skills and organizational change, he warned, companies could lose both the productivity benefits of AI and employees to competitors that adopt the technology more effectively.

Add as a preferred source on Google +