Tiwa York is a seasoned entrepreneur, investor, and advisor with over 20 years of experience in Southeast Asia’s tech ecosystem. He founded and led Kaidee.com, Thailand’s largest online marketplace, through its growth and successful acquisition in 2020. Now a Principal at Finno Efra Accelerator and Board Member at Seven Peaks Software, Tiwa supports startups across fintech, SaaS, foodtech, and proptech. Known for his hands-on approach and strategic insight, he continues to mentor founders and guide high-growth ventures from his base in Bangkok.
Earlier this month at Seven Peaks, over 200 attendees gathered for Automate or Be Automated, a high-impact session on the current state of AI. From the outset, it was clear this wasn’t just another tech talk: within minutes, speaker Tiwa York used AI to generate a complete strategic research report for a volunteer—live, unscripted, and unprepared. It was a vivid demonstration of how far AI capabilities have come, and how quickly they’re reshaping business strategy.
If there was a single takeaway from the evening, it’s this: 2025 marks a turning point. AI agents are no longer experimental. They’re becoming essential.
Tiwa opened with a data point that sparked reflection: while just one percent of knowledge workers qualify as AI experts, more than half believe they already understand AI well. The gap between perception and capability is stark and helps explain why some organizations are accelerating ahead while others lag behind.
"The cost of digital goods is trending toward zero, while their value keeps rising," Tiwa noted, underscoring the urgency of adopting AI not as a novelty but as an engine for scalable value creation.
A key theme of the evening was demystifying what today’s AI systems actually do. Large Language Models (LLMs) like GPT-4 and Claude don’t "think" or "understand" in a human sense—they predict, based on patterns in vast training datasets.
One live example drove this home. Consider how an LLM might complete this sentence: "I grabbed my keys and rushed to the ______."
Given a the context of "my sister was in labor, and I didn’t want to miss the birth,” it would write "hospital." But given the context of “The waves were perfect, and I didn’t want to miss my morning surf session,” it would write "beach."
The shift in output, driven entirely by context, illustrates how LLMs rely on correlation, not cognition.
Yet AI is evolving. Projects like Google’s AlphaProof and AlphaGeometry (for mathematical reasoning) mark a new frontier: AI systems that go beyond pattern matching to structured reasoning. This is a critical development for use cases that require accuracy and depth.
While public attention often centers on ChatGPT or image generation tools, the real transformation is happening with AI agents. These systems can plan, make decisions, and take action with minimal human input.
Case studies from the session made the business impact clear:
These are not pilot programs. They are active implementations with measurable outcomes across departments like legal, sales, and operations.
Another standout theme was "Vibe Coding." This approach to building software through natural language and intuitive tools is reducing reliance on traditional programming.
The ecosystem is rapidly maturing:
For entrepreneurs, this means dramatically lower development costs and faster time-to-market. Vision, not technical skill, is increasingly the barrier to launching powerful digital products.
Across case studies, a consistent pattern emerged among organizations successfully leveraging AI. Four practices stood out:
Another key insight: effective prompting makes a significant difference. Tiwa highlighted the PTCF framework (Persona, Task, Context, Format) as a best practice for getting useful outputs from AI systems.
Several examples underscored how AI is transforming strategic decision-making.
In both cases, early adopters are gaining operational advantages while creating intelligence loops that improve over time.
The evening concluded with a look ahead at reasoning-driven AI systems with tools that don’t just respond, but plan, solve, and act proactively.
One example was Manus.im, a platform combining multiple agent types (planners, validators, researchers) into one autonomous system. It recently raised $85 million at a $500 million valuation, a signal of investor confidence in next-gen AI that acts while users sleep.
These are not distant hypotheticals. Such systems are launching now, and they represent a shift from reactive tools to autonomous partners.
The core message was clear: businesses must either learn to harness automation or risk being sidelined by it. The leaders of the next decade will be those who don’t just use AI, but integrate it strategically, systemically, and creatively.
The tools are ready. The frameworks are emerging. The gains are measurable. The real question is: are you ready to lead in this new era?
Explore further: Access the full slide deck, case studies, and tools discussed during the event.
Need support? Our team helps startups and enterprises build AI-native products from concept through deployment. Let’s explore how intelligent systems can create real value for your business.
Tiwa York
Angel Investor - Seven Peaks Board Member
Tiwa York is a Bangkok-based entrepreneur and investor with over 20 years of experience. He founded Kaidee.com and now serves as a Board Member at Seven Peaks Software and is Principal at Finno Efra Accelerator, supporting startups across fintech, SaaS, foodtech, and proptech.