Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The digital illustration of the world’s luminous map with the AI text on several continents, which presents the global presence and integration of artificial intelligence.
Fotograzia | Moment | Gets the image
As artificial intelligence becomes more democratized, it is important for new economies to build their own “sovereign II”, – said CNBC discussions Western Conference East Tech In Bangkok, Thailand, Friday.
In general, the sovereign II refers to the country’s ability to control its own AI technologies, data and related infrastructure, providing strategic autonomy when satisfying its unique priorities and security needs.
However, this sovereignty lacked, according to the discussion of Kasima Tarnpipichi, the head of the AI SCB 10x strategy, the SCBX Group technology investment group. He noted that many of the most famous large language models are guided by companies such as anthropic and Openai, based on English.
“The way you think you interact with the world, which you are when you speak another language, may be completely different,” Tarnpipich said.
Thus, it is important for the countries to take on the property of AI systems, developing technology for certain languages, cultures and countries, not just translation into Anglo-models.
The participants of the discussion agreed that the digital savvy region of ASEAN with the total population of almost 700 million people, especially able to create their sovereign II. People under the age of 35 make up about 61% of the population, and about 125 000 new users Enable Internet access daily.
Given this context, Jeff Johnson, Head Director ASEAN AT Amazon The web service said, “I think it’s really important and we are really focused on how we can truly democratize access to Cloud and AI.”
Models with open source
According to the discussion participants, one of the key ways that countries can create their sovereign AI environments is the use of AI models with open source.
“There are many strange talents here, in South -East Asia and Thailand, especially. To be enthusiastic so as not to publicly access or develop an ecosystem, it feels ashamed,” said Thaarnpipithai SCB 10x.
To do with an open source is a way to create “collective energy” to help Thailand better compete in the II and push sovereignty in such a way that it is useful for the whole country, he added.
The open outcome usually applies to the software in which the source code is freely available, allowing anyone to view, change and redistribute it. LLM players, such as Chinese Deepseek and Llama Meta, advertise their models with both open source, albeit with some restrictions.
The appearance of more open source models offers companies and government more options compared to several closed models, according to Cecily NG, Vice President and General Manager ASEAN & Eight China at software suppliers.
AI experts have Previously told CNBC This open source AI has helped China increase AI adoption, it is better to develop its AI ecosystem and compete with the US
Access to computing
Prime Pavan, Vice President and CEO of Southeast Asia and Korea in Red Hat, said that until recently, localization II was concentrated in the language. The availability of sovereign access to AI models operating on local equipment and calculations is more important today, he added.
Participants in the discussion point out that for countries such as Thailand, the localization of the II may offer cloud computing companies with domestic activities. These include global hyperskalers such as AWS, Microsoft Unhappy and Favorable Cloud and sovereign players such as AIS Cloud and True IDC.
“We are here in Thailand and in Southeast Asia to support all industries, all enterprises of all forms and sizes, from the smallest startup to the largest enterprise,” said Johnson Aws.
He added that the economic model of cloud services makes it easier to “pay for what you use”, thereby reducing the entry barriers and makes it very easy to build models and applications.
In April, the UN Agency for Trade and Development stated in report It is forecast that AI will reach 4.8 trillion by 2033. Dollars. However, he warned that technology benefits remain highly concentrated, and nations are underway behind.
Among the recommendations of the UNCTAD International Community for Inclusive Growth Management was the AI AI Infrastructure, the use of AI models and open source initiatives to exchange AI knowledge and resources.
