Listen to this article:
ATxSummit 2026: From ideas to deployment
The ATxSummit, organized by the Singaporean Infocomm Media Development Authority (IMDA) together with Informa, was held this year in the spirit of practical AI application. The motto of the year was "from concepts to concrete results" — and the partnership announcements matched. The event took place on May 19–20, 2026 at Singapore's Gardens by the Bay, attended by President Tharman Shanmugaratnam, Minister Josephine Teo, as well as key figures from the global AI scene such as OpenAI's Chief Revenue Officer Denise Dresser and NVIDIA's Chief Scientist William Dally.
Singapore has also established the National AI Council under the leadership of Prime Minister Lawrence Wong and defined four priority sectors for national AI missions: connectivity and transport, advanced manufacturing, healthcare, and finance. These four areas together account for over 40% of Singapore's GDP.
OpenAI gives Singapore over 300 million dollars — and wants to turn it into a laboratory for the world
The biggest announcement of the summit came from OpenAI, which signed a memorandum of understanding with the Ministry of Digital Development and Information (MDDI). The company committed to investing over 300 million Singapore dollars (approximately 234 million USD, or about 5.2 billion CZK) in Singapore's AI ecosystem.
The partnership called "OpenAI for Singapore" rests on three pillars: supporting applied AI research, building a talent base, and making AI accessible to citizens, businesses, and public administration. For Czech readers, the key takeaway is that OpenAI is building a model that could later be replicated in Europe as well — but for now, Singapore serves as a testing ground for faster and less regulated deployment.
NVIDIA opens a research lab: embodied AI and energy-efficient computing
NVIDIA announced the creation of an AI Research Lab directly in Singapore. This is the company's second research center in the Asia-Pacific region, focusing on two directions: embodied AI — intelligent systems capable of perceiving, reasoning, and acting in the physical world — and efficient AI computing, which optimizes models and infrastructure with the goal of reducing computational costs and energy consumption.
The lab will collaborate with university researchers, industry partners, and government agencies. The primary application domain is manufacturing — Singapore already ranks among the countries with the highest density of industrial robots in the world (roughly five times the global average). For Europe, and especially for Czech industry with its strong manufacturing tradition, this is a signal of where the intersection of physical robotics and artificial intelligence is heading.
Punggol Digital District: The first public testing ground for robots
However, the most concrete project is undoubtedly the Physical AI Testbed in the Punggol Digital District. Singapore's IMDA, the JTC agency, and the Singapore Institute of Technology (SIT) will launch the first test zone this year for deploying robots and embodied AI systems in a regular mixed public environment.
What makes it unique is that it will be possible to test multiple operators and multiple types of robots simultaneously — something no other public testing ground in the world offers. The first industry partners are:
- Certis — security patrolling
- DHL — parcel delivery
- Grab — food delivery
- QuikBot — robotic services
Additionally, Singapore's Land Transport Authority (LTA) has granted a district-level exemption under the Active Mobility Act, allowing robots to legally operate on the streets alongside people. The goal is to test safety mechanisms, regulatory aspects, and infrastructure requirements for future large-scale deployment of robots.
What this means for Europe and the Czech Republic
Singapore's approach to AI regulation and testing offers an interesting contrast to the European model. While the EU bets on a comprehensive legal framework in the form of the AI Act and every real-world test goes through lengthy approval processes, Singapore is taking the path of regulatory sandboxes and district-level exemptions, allowing companies to quickly test in real operations.
For Czech companies and startups, this means two things. First, if they want to test robotic or embodied AI systems in a real environment with minimal regulatory burden, Singapore is becoming an attractive destination. Second, if Europe does not learn to create similarly flexible test zones, European companies in the field of physical AI will fall behind Asian competitors.
The Czech Republic is not lagging behind in terms of AI infrastructure — the Czech AI Factory in Ostrava has become part of the European network and Czech companies like Ecomail are already actively working with AI models. However, what is missing is precisely the ability to test robotic systems in regular operation without excessive administrative burden.
Agentic AI gets new rules
The summit also brought an update to the Model AI Governance Framework for Agentic AI, whose version 1.5 now includes over ten new case studies from more than fifty organizations, including DBS, Google, PwC, and Tencent. Singapore, together with Google and the Cyber Security Agency (CSA), also published a practical white paper for deploying AI agents in public administration.
This combination — rapid real-world testing plus thoughtful governance — is proving to be a pragmatic path from which the EU could take inspiration. Singapore does not compromise between speed and safety, but builds both in parallel.
Temus AI Foundry and other investments in talent
The company Temus, founded by the Singaporean state fund Temasek, launched the AI Foundry with support from Digital Industry Singapore (DISG), which will hire 50 specialists — from AI architects through data scientists to ML engineers. It will focus on financial services and precision healthcare. Singapore also launched the National AI Impact Programme with the aim of bringing AI to 10,000 small and medium enterprises.
For context: approximately 70 AI Centers of Excellence already operate in Singapore. By comparison, the entire Czech Republic — with a similar population — has orders of magnitude fewer such centers.
Are the technologies presented at ATxSummit 2026 available to companies outside Singapore as well?
Yes, all announced partnerships have an international dimension. Research results from the NVIDIA lab will be published openly, and OpenAI is building a model in Singapore that it plans to replicate in other countries. The Punggol Digital District is also open to foreign companies — but they must have at least a local presence in Singapore.
How does Singapore's approach to regulating agentic AI differ from the European AI Act?
Singapore uses the principle of "regulatory sandboxes" — instead of blanket bans or approvals, it creates temporary exemptions within a specific area (such as Punggol), where companies test AI in real operations under the regulator's supervision. The EU, by contrast, builds on horizontal legislation that classifies AI systems by risk and requires certification before deployment. The Singaporean model is faster, the European one more thorough — an ideal scenario would be a combination of both.
Could Singapore serve as inspiration for the Czech AI strategy?
Absolutely, especially in the area of testing robotic systems. The Czech Republic has a strong industrial base and a high density of industrial robots (similar to Singapore), but it lacks a legislative framework for testing autonomous systems in public spaces. The district-level exemption model used by Singapore for Punggol could also work in Czech industrial zones or university campuses.
