SK Group unveils AI flywheel strategy to link chips and energy
The South Korean conglomerate plans to create a self-reinforcing cycle where semiconductor growth drives energy demand and new AI services.

The short version
- SK Group aims to link its semiconductor, energy, and infrastructure businesses into an 'AI flywheel' model.
- The strategy relies on AI chips driving the expansion of AI factories, which in turn increases demand for electricity.
- SK plans to enter the token economy to trade computing resources and AI services.
- The group is currently using AI and digital twins to upgrade 'smart' factories and automate network management.
Why it matters
This move signals a shift from being a diversified conglomerate to a vertically integrated AI powerhouse. For the reader, it highlights how industrial giants are trying to solve the massive power and hardware bottlenecks currently facing the AI boom.
The AI flywheel concept
SK Group Chairman Chey Tae-won recently introduced a new strategic framework called the 'AI flywheel' during a seminar in Icheon, South Korea. This strategy is designed to connect the group's existing strengths in semiconductors, infrastructure, and energy with emerging artificial intelligence services. The goal is to move the conglomerate beyond its traditional business models to become a dedicated AI company.
The core idea behind a flywheel is a self-reinforcing cycle where each part of the business accelerates the growth of the others. In this specific model, the group intends for its AI chips to provide the necessary hardware for expanding AI factories. As these factories grow, they will create a massive, consistent demand for electricity and energy infrastructure, which are also core parts of SK's portfolio.
Chairman Chey Tae-won emphasized that no single company can succeed in the AI era by working in isolation. He called on the group's subsidiaries to seek open cooperation with global partners to build these capabilities. By linking these different sectors, the group hopes to create a loop where growth in one area, such as semiconductors, directly fuels investment and innovation in another, like energy services.
Expanding into models and tokens
Beyond just hardware and energy, SK Group plans to deepen its involvement in the software side of the industry. The company intends to expand its presence in AI models and specific applications. This means they are looking to move up the value chain from providing the raw components to providing the actual intelligence that powers modern software.
One of the more specific components of this strategy involves participating in the so-called token economy. In this economic model, computing resources and AI services can be traded as digital tokens. By entering this space, SK aims to create a marketplace where their infrastructure and AI capabilities can be traded more fluidly with other entities. This could allow for a more dynamic way to monetize their high-powered computing assets.
To support these goals, the group is also focusing on internal efficiency through a project called Portfolio Rebalancing 2.0. This initiative is designed to optimize the group's operations by identifying and strengthening core business areas while cutting out inefficient ones. The leadership wants to ensure that the company's internal structure is lean enough to support the heavy demands of an AI-first transition.
Practical AI in current operations
While the flywheel is a long-term strategy, SK is already implementing AI tools across its various subsidiaries to improve productivity. SK hynix, the group's semiconductor arm, is currently upgrading its 'smart' factories. These facilities use AI and digital twins—virtual replicas of physical systems—to improve manufacturing efficiency and monitor production in real time.
Other divisions are also integrating these technologies into their daily workflows. SK Innovation has introduced an ontology-based process map at its Ulsan CLX complex. This system combines data from equipment and operations to improve how they manage production performance. By mapping out these processes, the company can better understand how different variables affect their output.
The group's telecommunications and gas divisions are also seeing changes. SK Telecom is using AI to automate the management of its networks and services, reducing the need for manual oversight. Meanwhile, SK Gas is employing AI to analyze complex market conditions and organize the knowledge gathered from their previous operations. These examples show that the group is using AI to refine its existing business before fully pivoting to new services.
The role of energy and chips
A major hurdle for the current AI boom is the sheer amount of power required to run large-scale models. SK Group's strategy directly addresses this by linking their energy infrastructure with their semiconductor production. By controlling the power supply and the chips that consume it, the group aims to create a more stable environment for AI development.
The group believes that the expansion of AI factories will be a primary driver for electricity demand. This creates a natural synergy where the growth of the AI industry provides a guaranteed customer base for SK's energy services. In turn, the profits from energy and infrastructure can be reinvested into more advanced semiconductor research and innovation.
This vertical integration is intended to create a more resilient business model. Instead of just selling chips to other companies, SK wants to be part of the entire ecosystem. By owning the energy, the hardware, and the services, they hope to capture more value from the AI lifecycle. As reported by Korea Herald, this move is intended to strengthen the group's capabilities and accelerate their overall development speed.
Future outlook and risks
The transition to an AI-centric company is not without its challenges. During the seminar, executives discussed both the new opportunities and the potential risks associated with this massive shift. The group is aware that moving into the token economy and AI model development requires significant capital and a high level of technical coordination with global partners.
To manage these risks, the group is focusing on the Portfolio Rebalancing 2.0 initiative. This plan is meant to ensure that the company does not overextend itself by trying to do everything at once. By optimizing their current operations, they aim to create a solid financial foundation that can support the heavy investments needed for the AI flywheel to take off.
The ultimate goal is to transform the group into a company where AI is not just a tool, but the core of the business. If successful, the flywheel will create a cycle of growth where each success in one sector feeds the next. For now, the group is focused on the practical steps of upgrading factories and automating networks to prepare for this larger transformation.
Who is affected
The primary stakeholders in this shift are the subsidiaries of the SK Group, which will see their business models reimagined. Executives from companies like SK hynix, SK Innovation, and SK Telecom will be responsible for integrating these new AI tools into their specific industries. This means a shift in how these companies manage production, networks, and market analysis.
Global partners will also be affected, as Chairman Chey Tae-won explicitly called for open cooperation. Companies that provide complementary technologies or services may find new opportunities to partner with SK as it builds out its AI infrastructure. This could lead to new types of contracts and joint ventures in the semiconductor and energy sectors.
Finally, the broader tech industry may see a shift in how AI infrastructure is sold and traded. By entering the token economy, SK Group could influence how computing resources are distributed. For the general public, these changes might eventually manifest as more efficient industrial processes and a wider variety of AI-driven services provided by the group.
Next steps for the group
The immediate next step for SK Group is the continued execution of the Portfolio Rebalancing 2.0 initiative. This will involve identifying which parts of the current business are most efficient and which need to be streamlined to support the AI flywheel. The group will be looking for ways to maximize productivity and profitability across all its holdings.
Simultaneously, the group will continue to expand the practical use of AI in its existing operations. This includes the ongoing upgrades to smart factories and the automation of network management. These projects serve as the testing ground for the larger AI services the group hopes to launch in the coming years.
Looking further ahead, the success of the strategy will depend on the group's ability to secure global partnerships. SK will need to navigate the complexities of the token economy and the high-stakes world of AI model development. The group's ability to turn its energy and chip strengths into a self-sustaining AI cycle will be the key metric for its success by 2026.
Summary of key moves
SK Group is moving toward a model where their various business units work in a cycle to support AI growth. This involves using their chips to build AI factories, which then creates a demand for their energy and infrastructure. This cycle is intended to create a self-sustaining loop of investment and innovation.
The group is also looking to move into the software and economic side of AI. They plan to develop AI models and participate in a token economy where computing power can be traded. This represents a significant expansion from their traditional role as a hardware and industrial provider.
To make this happen, SK is currently using AI to improve its own internal processes. From digital twins in factories to automated network management, the group is preparing its internal infrastructure for a major shift. By balancing these internal improvements with a new external strategy, SK aims to become a leading AI company by 2026.
Technical terms explained
The 'AI flywheel' refers to a business model where different parts of a company reinforce each other's growth. For example, more customers for one product can lead to more data, which improves the product for everyone else, creating a cycle of continuous improvement.
A 'digital twin' is a virtual model of a physical object or system. In manufacturing, it allows companies to simulate how a factory will behave in a computer program before making changes in the real world. This helps in identifying problems and improving efficiency without risking actual production.
The 'token economy' is a system where assets, such as computing power or AI services, are represented by digital tokens. These tokens can be traded or sold on a blockchain or other digital ledger. This allows for a more flexible and automated way to trade resources between different companies and users.
Strategic summary
SK Group's strategy is a response to the massive infrastructure needs of the current AI era. By linking energy, chips, and services, they are trying to solve the power and hardware bottlenecks that currently limit AI growth. This vertical integration is a way to ensure they have the resources needed to compete with larger tech giants.
The group is also trying to capture more value by moving into the software and token markets. Instead of just providing the 'pipes' for the AI industry, they want to own the 'water' as well. This shift from a hardware-centric to a service-centric model is a major pivot for a traditional conglomerate.
Ultimately, the success of this plan depends on the group's ability to execute on its internal optimizations. If they can make their current operations more profitable and efficient, they will have the capital and the focus needed to build out the AI flywheel. The goal is to create a company where every business unit contributes to a larger, AI-driven ecosystem.
The roadmap to 2026
The seminar in Icheon served as a launchpad for a multi-year plan. The group has set a clear focus on the next few years to transition its core identity. This involves not just adding AI to their existing products, but making AI the central theme of how the entire group operates and generates revenue.
The group will be monitoring several key metrics as they move forward. These include the efficiency gains in their 'smart' factories, the success of their process maps in the Ulsan complex, and the growth of their AI-driven network management. These are the building blocks for the larger flywheel strategy.
As SK Group moves toward 2026, the industry will be watching how they manage their global partnerships. Success will require a balance of internal innovation and external collaboration. If they can successfully link their energy, chips, and services, they could become a dominant force in the global AI infrastructure market.
Operational changes
The transition to an AI-first company involves significant changes in how SK Group's subsidiaries operate. SK hynix is moving toward a model where digital twins and AI guide manufacturing decisions. This reduces human error and allows for faster adjustments to production schedules based on real-time data.
SK Innovation is using ontology-based process maps to organize their operational knowledge. This means they are creating a structured way for the company to 'understand' its own data. By mapping out how equipment and operations interact, they can identify bottlenecks and improve performance across their complexes.
SK Telecom and SK Gas are also seeing changes in their daily management. Automation in network management allows for faster response times to issues, while AI-driven market analysis for gas helps the company stay ahead of price fluctuations. These changes show that the group is already integrating AI into the core of its business operations.
The role of semiconductors
Semiconductors are the foundation of the SK Group's AI flywheel. The group's AI chips are intended to be the primary hardware that powers the expansion of AI factories. By producing the chips, SK can ensure they have the necessary components to build out the infrastructure they need.
This creates a direct link between the semiconductor business and the energy business. As more AI chips are produced and used, the demand for electricity increases. Because SK also provides energy infrastructure, they can capture a larger share of the profits from the AI boom than a company that only makes chips.
This strategy aims to create a more stable and profitable ecosystem for the group. By controlling both the supply of chips and the demand for energy, they can better manage the risks and rewards of the AI industry. It is a move toward vertical integration that seeks to maximize the value of every part of their portfolio.
Energy and infrastructure
Energy is the fuel that powers the AI revolution. SK Group recognizes that as AI factories expand, the demand for reliable and massive amounts of electricity will skyrocket. By leveraging their existing infrastructure and energy businesses, they can position themselves as a primary provider for the AI industry.
The group's strategy suggests that growth in the AI sector will directly spur investment in their energy services. This means that as AI companies build more data centers and factories, SK's energy division will see increased demand. This creates a reliable revenue stream that can be used to fund further innovation in other areas.
This synergy is a key part of the 'flywheel' concept. The more AI infrastructure is built, the more energy is needed, which fuels the growth of SK's energy business. This growth, in turn, provides the resources to develop more advanced chips and AI services, completing the cycle and driving the group's overall expansion.
AI services and the token economy
SK Group is looking to move beyond just providing the hardware and energy for AI. They also want to be a major player in the software and service side of the industry. This includes developing their own AI models and applications that can be used by other businesses and consumers.
A major part of this plan is participating in the token economy. This is a system where computing resources and AI services can be traded as digital tokens. By entering this market, SK can create a platform where their high-powered computing assets can be sold and traded more easily with global partners.
This move into the token economy could open up new revenue streams for the group. It allows them to monetize their infrastructure in a more dynamic way, potentially reaching a wider audience of developers and companies. It also aligns with the broader trend of using blockchain and tokenization to manage complex digital resources.
Global cooperation and partnerships
Chairman Chey Tae-won has emphasized that the group cannot succeed in the AI era alone. He has called for open cooperation with global partners to strengthen the group's capabilities. This recognizes that the AI landscape is vast and requires a wide range of expertise and technologies that no single company can provide on its own.
By seeking global partnerships, SK Group aims to accelerate its development and stay competitive. This could involve collaborating with other tech giants, research institutions, or specialized AI startups. These partnerships will be essential for building out the complex infrastructure required for the AI flywheel to function.
This open approach to cooperation is a key part of the group's strategy. It allows them to leverage the strengths of others while contributing their own expertise in semiconductors and energy. By working with global partners, SK can more quickly scale its AI services and reach a broader international market.
Internal optimization and rebalancing
To support this massive strategic shift, SK Group is focusing on internal efficiency. The Portfolio Rebalancing 2.0 initiative is a key part of this effort. It is designed to optimize the group's operations by identifying and strengthening core business areas while cutting out inefficient ones.
This rebalancing is necessary to ensure that the group has the resources and focus needed to succeed in the AI market. By streamlining their operations, they can allocate more capital and talent to their high-growth areas like semiconductors and AI services. It is a way of making the company leaner and more agile.
The group is already seeing the results of these efforts in its various subsidiaries. From upgrading 'smart' factories to automating network management, the focus is on using AI to improve productivity and profitability. These internal improvements provide the foundation upon which the larger AI flywheel strategy will be built.
Sources and links
- ReportingUA.NEWS: the article we sourced
Liked it? Say why
Add a line for other readers.
No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.
Listen to the AI Wire headlines
Listen to the AI WireListen to our daily AI shows every morning: on your commute, in the shower, wherever you are
Listen to the daily showsFirst reported by UA.NEWS · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.
