AI can make market research faster, improve buyer communication and streamline everyday business processes. But using AI is the easy part. Using it responsibly, turning it into practical solutions and transferring that knowledge to SMEs? That takes skills.
That is exactly what sequa set out to strengthen through its intervention under the GIZ programme EU-German Global Access and Trade Expertise. Globally Cool’s Onno delivered a five-day Train-the-Trainer programme on AI Solutions for Export Promotion in Phnom Penh from 31 August to 4 September 2026. The goal was clear: equip Cambodian consultants, business-support professionals and other SME-support actors with practical AI skills they can use themselves, and pass on to companies across Cambodia.
Five days. One practical AI journey.
We designed the programme around one principle: learning by doing. Rather than spending five days talking about artificial intelligence, participants worked with it. They researched markets, challenged AI-generated answers, created buyer communication, built AI-supported workflows and ultimately developed their own training and advisory services.
And throughout the programme, one message remained central: AI is only useful when reliable sources, verification and human judgement stay firmly in the loop.
Starting with responsible AI
Participants explored how AI can produce unsupported or overly confident answers and learned how to recognise hallucinations and assumptions. They compared AI assistants, practised verification and developed structured Golden Prompts based on role, context, task, requirements and output format. They also experimented with NotebookLM to turn trusted source material into learning content.
Why start here? Because good AI for export promotion is not about asking a chatbot a question and copying the answer. It is about knowing how to brief AI properly, challenge its output and decide what can, and cannot, be trusted.
Putting AI to work for export market research
Once the foundations were in place, we moved into one of AI’s most useful applications for exporters: market research. Participants used Deep Research and complementary sources to explore potential export markets, buyer needs, channels, competition, market-entry requirements and trends.
But there was a catch: the first AI answer was never automatically the final answer. Participants had to cross-check important findings before turning the research into buyer-focused value propositions, written communication and visuals. This evidence-first approach makes AI faster without sacrificing the quality of the market intelligence behind an export decision.
And it clearly resonated. Participants rated their ability to use AI for export market research and buyer communication at 4.7 out of 5, the programme’s highest self-assessed learning score.
From one-off prompts to reusable AI workflows
AI becomes much more interesting when it stops being a one-off tool and starts supporting repeatable business processes. Participants designed and tested AI assistant concepts using approved knowledge, rules and guardrails. They looked for repetitive export-related tasks where AI could save time or improve consistency and used structured ChatGPT Work processes to create business communication and digital content.
Just as importantly, they reviewed their outputs from the perspective of an international buyer. The objective was not to automate human judgement away. It was to understand where AI can take over repetitive work while people remain responsible for the decisions that matter.
Turning AI skills into services SMEs can actually use
Knowing how to use AI is one thing. Building a service around it is another. On day four, we changed the question from “Which AI tool can I use?” to “Which customer problem can I solve?”
Participants researched the Cambodian market, defined target customers and developed service propositions, Business Plans and 90-day launch actions. They also created Trainer Manuals, exercises and supporting materials that could help them transfer their new skills to SMEs, colleagues and other professionals.
This step was essential to the Train-the-Trainer approach. The aim was never just to create a group of better AI users. It was to build local capacity for practical, responsible and potentially sustainable SME export support in Cambodia.
The Final Challenge: time to prove it
In the end, four teams had to bring everything together and present a Business Plan, training or advisory concept and evidence showing how AI had contributed to their solution.
Across the four concepts, participants combined tools including ChatGPT, Claude, Perplexity, NotebookLM, Synthesia and Deep Research. Their outputs ranged from market research and business plans to Trainer Manuals, infographics, buyer communication and videos. The impressive part was not how many AI tools they used. It was how purposefully they used them.
From AI skills to stronger export services
By the end of the programme, participants had moved well beyond experimenting with prompts. They were using AI to conduct export market research, improve buyer communication, create repeatable workflows, design training and build services around real SME challenges. That creates a strong starting point for locally anchored AI-enabled export support in Cambodia. Now comes the most exciting part: turning those ideas into real services, testing them with Cambodian SMEs and learning what creates the most value in practice.
Want to strengthen your organisation’s export services with practical AI solutions and hands-on capacity building? Get in touch with us.