The next evolution of artificial intelligence is moving beyond content creation into autonomous action, a critical shift from “Generative AI” to “Agentic AI” that will fundamentally reshape the business models of Software-as-a-Service (SaaS) startups. This transformation demands a new strategic approach, focusing on flexibility, trust, and measurable value to stay competitive.
These insights were shared by Nicha Suebwonglee, Venture Capital Business Development Manager for Thailand & ASEAN at Amazon Web Services (AWS), who laid out a roadmap for startups to navigate this emerging landscape.
The transition comes as AI solidifies its role as a core business function, not just a technological trend. “The question is no longer ‘if’ a business should use AI, but ‘what’s next and how do we leverage it?’” Suebwonglee explained. This reality is backed by compelling data: in 2024, 72% of global organizations had already experimented with generative AI, with 40% confirming tangible productivity gains. Looking ahead to 2025, an estimated 60% of companies are expected to appoint a Chief AI Officer, and 45% have designated AI as a top budget priority.
Amid this widespread adoption, the technological frontier is advancing toward Agentic AI—autonomous systems capable of reasoning, planning, and executing complex tasks on behalf of a user with minimal human oversight.
“Think of the evolution from single-task machine learning to the generative models we chat with today. The next step is an autonomous agent,” Suebwonglee said. “Formally, it’s an autonomous software system that uses AI to achieve complex goals. You don’t just ask it to write something; you assign it a mission, like ‘monitor all incoming bug reports and autonomously resolve them.’”
This requires a combination of knowledge, tools, autonomy, decision-making ability, and the capacity to reason and reflect on its actions.
To harness this power, Suebwonglee outlined three pillars for success.
First is Choice and Flexibility. “The AI model landscape is changing daily,” she noted. “Instead of betting on a single model, startups need a flexible architecture to test, evaluate, and swap models to find the most cost-effective solution for each use case.” While access to models becomes commoditized, she stressed that a startup’s proprietary, high-quality data becomes its single most important differentiator.
The second pillar is building Trustworthy AI, a non-negotiable for enterprise clients. This extends beyond robust security and privacy—which includes access controls, auditability, and data protection—to encompass Responsible AI. “This means implementing strong governance, clear testing standards, and guardrails to prevent harmful or inappropriate outputs, all while ensuring transparency and accountability,” Suebwonglee advised.
Finally, startups must focus on Maximizing Value. “Start with a clear business problem, not the technology,” she urged. This involves a systematic process of evaluating a use case for its data readiness and compliance needs, assessing its potential ROI and strategic importance, and prioritizing projects that balance technical feasibility with the highest business impact, such as using AI for developer code generation or creating hyper-personalized customer experiences.
Looking ahead, the rise of Agentic AI is predicted to trigger a complete paradigm shift for the SaaS industry. The core offering will evolve from providing a tool (Software-as-a-Service) to delivering a direct Outcome-as-a-Service. Consequently, pricing models will move away from simple monthly subscriptions toward value-based pricing or even an “AI Agent-for-Hire” model, where capabilities are billed like a human consultant.
“Perhaps the most radical change will be in customer acquisition,” Suebwonglee concluded. “As your customers adopt their own AI agents, your product must be discoverable and interoperable not just with humans, but with other AIs. The future sales process may very well involve your AI agent communicating with your customer’s AI agent.”
True Corp. reports net profit of THB 2.0 Bn for Q2/2568
Thailand’s inclusive finance push: Using AI and data to Fuel growth
