Singapore is looking to make a global mark in artificial intelligence (AI) with the issuance of international standards for large language model (LLM) testing and investment in quantum computing and new data center capacity.
Deputy Prime Minister Heng Swee Keat said at the Asia Tech x Singapore 2024 summit last week that quantum has the potential to unlock new value, where higher processing capacity can be exploited in fields areas such as simulation of complex molecules for drug discovery.
Also: Innovative AI may be creating more jobs than it saves
He added that quantum computing could also have synergies with AI, such as in improving the efficiency of developing and training advanced AI models. This development, in turn, could further spur innovations in the fields of deep learning, natural language processing, and computer vision.
However, there are still challenges to be solved on the quantum side, including cryogenic cooling and error correction requirements, Heng said. He noted that researchers around the world are evaluating different approaches to achieve scale and enable quantum computing to become commercially viable.
Also: Last year’s Rote automation was the same: AI pushed more intelligence into software development
Singapore wants to solve these problems challenge with the National Quantum Strategy, along with an investment of nearly 300 million Singapore dollars (221.99 million USD). This cash is on one previously committed S$96.6 million The new investment is earmarked for five years, until 2030, to enhance the country’s position as a leading hub in the development and deployment of quantum technology, Heng said.
This roadmap focuses on four areas, including initiatives in quantum research, such as quantum communications and security and quantum processors, as well as a scholarship program that aims to produce 100 doctoral and 100 master’s-level graduates over the next five years, he said.
Singapore is making efforts to build the ability to design and develop quantum processors. This work will include research on qubit technologyincludes photonic networks, neutral atoms and superconducting circuits.
ZDNET understands that Singapore’s goal is to be ready for the first prototype in the next three years and to scale up production within five years.
The government in 2022 announced a three-year initiative to Build a quantum secure network that it hopes to introduce “agile crypto connectivity” and facilitate trials with both public and private entities. The initiative also includes a quantum security lab for vulnerability research.
Lay the foundation for green data centers
Singapore also launched last week Green data center roadmap to chart “digital sustainability and chart a green growth path” for such facilities, supporting AI and computing developments.
The country has more than 1.4 gigawatts of data center capacity and is home to more than 70 cloud, enterprise and co-location data centers.
Singapore is aiming to add at least 300 MW of electricity additional data center capacity and another 200 megawatts through green energy deployment, Janil Puthucheary, Senior Minister of Communications and Information, said at the summit.
Efforts will be made to increase quality through both hardware and software, Puthucheary said, pointing to technologies that maximize energy efficiency and capacity as well as green software tools.
He added that improving data center efficiency also involves greening software, which can reduce the carbon footprint of applications.
He said the focus would be placed on data centers to accelerate the use of green energy, with the government providing support through grants and incentives to switch to energy-efficient IT equipment. quantity. Additionally, the Infocomm Media Development Authority (IMDA) will work with PUB to help data centers increase their water use efficiency (WUE) to 2.0 cubic meters or less per megawatt hour, an increase from with an average 2021 WUE of 2.2 cubic meters.
Also: Agile development can unleash the power of creative AI – here’s how
IMDA will jointly develop standards and certifications with industry partners to promote the development and operation of data centers with a power utilization efficiency (PUE) of 1.3 or less.
Additionally, the BCA-IMDA Green Mark for data centers will be refreshed later this year to raise the bar for energy efficiency in data centers. IMDA will also introduce standards for IT equipment energy efficiency and liquid cooling by 2025, aimed at driving the adoption of these technologies in Singapore.
The green data center roadmap outlines plans to reduce energy use for air cooling by increase operating temperature through IMDA Tropical DC Method.
According to the government agency, data centers can save energy by 2% to 5% for every 1°C increase in operating temperature.
It also points to simulations showing that existing data centers could reduce the energy consumption of supporting infrastructure by 50%, by making energy-efficient improvements and upgrades to critical devices. important, such as refrigeration plants and uninterruptible power supplies.
“We aim to upgrade all data centers in Singapore to achieve PUE below 1.3 with 100% IT load over the next 10 years,” IMDA said. “This gives existing data centers enough time to plan for upgrades.”
Today’s technology industry emits about 1.5% to 4% of global greenhouse gas emissions, Heng noted, and this number is expected to increase as the use of AI expands along with storage needs. and data processing.
Also: 3 Ways to Accelerate General AI Deployment and Optimization
He said the technologies driving the country’s digital economy, such as cloud and AI, drive demand for powerful and energy-intensive computing.
“Data centers are at the heart of such activities and require large amounts of energy for processing and cooling,” he said. Therefore, greening IT, especially data centers, is critical in a digital and carbon-constrained world.”
“There is a need to balance the economic and social benefits of digital applications with the environmental impacts of emissions,” he said. Also note that Singapore has committed to achieving its zero emissions target by 2050.
“The [green data center] Puthucheary explains that the roadmap sets out low-carbon energy sources that data centers could explore, including bioenergy, fuel cells capable of carbon capture, low-carbon hydrogen and ammonia to start. . These roads are in Singapore.”
Charting global testing standards for AI models
Meanwhile, the country wants to lead the way by introducing standards for large language model (LLM) testing, developed through partnerships with global organizations such as MLCommons, IBM and Singtel .
Named Project Moonshot, the LLM testing tool provides benchmarking, red-grouping, and testing baselines to help developers and organizations mitigate the risks associated with LLM deployments.
Also: Generative AI is the technology that IT feels the most pressure to exploit
Unfenced LLMs can reinforce biases and create harmful content, causing unintended consequences. “IMDA is looking to establish guardrails to manage risks while creating space for innovation,” the government agency said.
“It is important to adopt an agile, test and iterate approach to address key risks during model development and use. Project Moonshot provides intuitive results, so try Experiments demonstrate the quality and safety of a model or application in a way that is easy to understand, even for non-technical users.”
This testing tool provides a five-tiered scoring system, where each completed scoring sheet places the application on a scale. The author of each of these scoring tables can define the score limits.
AI verification organization and MLCommons jointly develop testing LLM standards. The latter is an open engineering consortium backed by Qualcomm, Google, Intel and NVIDIA and recognized by the US National Institute of Science and Technology under the AI Safety Consortium. AI Verify Foundation is a Singaporean non-profit organization focused on developing AI testing tools.
Also: The AI business is booming: ChatGPT Enterprise now has more than 600,000 users
Moonshot Project currently available as open beta.
IMDA says it is working with companies such as Anthropic to develop a practical guide to multilingual and cross-cultural collaboration for LLMs. This guide is expected to be released later this year for use globally.
