Conference Presentation, Panel
Smart Finance: The Intersection of AI, Technology, Talent, and Transformation | Asia Summit 2025
- The Singapore government is actively nudging businesses to scale and commercialize AI in 2025, supported by $100 million in grants for smaller companies to move from proof-of-concept to commercialization and a $150 million subsidy initiative focused on education and talent.
- Global AI total spend is projected to reach $500 billion annually across power, chipsets, data centers, and language models, with potential annual value creation estimated at $4 trillion to $6 trillion upon full realization.
- Despite 92% of companies planning increased AI investment, only 1% have realized bottom-line impact so far, while 95% of current AI initiatives lack a payback period.
- The industry is transitioning from foundational investment to applied AI and agentic systems, which shift from passive tools to proactive, autonomous problem-solving capabilities for multi-step tasks.
- T-Mobile achieved a $10 billion profitability increase in one year by becoming an early adopter that redefined workflows, illustrating that companies capable of using AI will detach from competition.
- Organizations currently face "pilot purgatory" where experimentation fails to scale, with winners needing to rewire organizations, transform workflows, and implement CEO-led change management to drive EBITDA impact.
- The strategic distance between AI winners and laggards will widen rapidly, as enterprises and countries failing to adapt risk obsolescence similar to Nokia, Blackberry, and Motorola during the wireless transition.
- 30% to 40% of jobs will be fundamentally different in the next three to five years, creating a moral responsibility for leaders to reskill and upskill employees rather than simply replacing them.
- Human adoption barriers remain the primary impediment to AI success, followed by data, governance, security, and partner issues, with people cited as the biggest challenge alongside the need to turn experimenters into accelerators.
- While agentic systems offer significant efficiency gains, such as reducing healthcare costs from $200–$220 to under $20 via improved diagnostic speed and accuracy, they require specific guardrails to prevent objective function drift and unintended side effects.
- Risk management must address real concerns, as 47% of companies experienced incidents like IP misuse, data privacy breaches, or inaccurate insights in the past 12 months.
- Leading organizations are piloting "managers of agents" roles to ensure human oversight remains critical, anticipating charts where agents and humans coexist as collaborators.
- Southeast Asian businesses must embrace the duality of US and Chinese technology stacks to create strategic flexibility and interoperability across different jurisdictions and regulatory landscapes.
- The US maintains an edge in semiconductors, while China has closed and surpassed gaps in language models, with projections indicating China will surpass American companies in AI in many cases within certain years.
- A fundamental reset is required for organizations unable to adapt, exposing leaders who cannot drive deep cultural change or redesign workflows without top-down executive involvement.
- Future selection criteria for professionals, such as stock pickers, will likely shift toward those who effectively leverage AI tools for insight and advisory rather than raw individual capability.
- WorldQuant and similar entities demonstrate that success involves hiring top global talent, such as 1,100 researchers, and providing voluntary workshops to ensure deployment of deep learning and neural networks, rather than replacing humans.