Has AI created a "bubble"? Will power shortages hinder AI development... A special interview with Chen Lan, Partner at Deloitte China.

Every Reporter | Zhang Huaishui Every Editor | Huang Bowen

On March 24, global attention once again gathered in Boao, Hainan.

On that morning, the Boao Forum for Asia announced two flagship reports for its 2026 annual meeting, namely the “2026 Annual Report on Asian Economic Outlook and Integration Process” (hereinafter referred to as the “Report”) and the “2026 Annual Report on Sustainable Development in Asia and the World.”

The “Report” pointed out that as the global focus of artificial intelligence development shifts from Europe and America to Asia, Asian economies are transforming from AI (artificial intelligence) followers to leaders, leveraging their massive digital population, rich application scenarios, and systematic policies to reshape the global AI innovation order.

With the global surge in AI enthusiasm, has the AI industry already generated an investment “bubble”? In the context of rising global electricity prices, will AI development face serious obstacles? Regarding these issues, a reporter from “Daily Economic News” interviewed Chen Lan, the partner and research director of Deloitte China, on-site at the annual meeting.

Ms. Chen Lan is one of the key interpreters of the flagship reports for the Boao Forum for Asia 2026 annual meeting and has over twenty years of experience in research related to new retail and the digital economy. She has been responsible for and participated in projects commissioned and delivered to national ministries such as the Ministry of Industry and Information Technology and the Ministry of Commerce.

Boao Forum for Asia 2026 Annual Meeting Image source: Provided by the organizer

Current AI applications continue to create substantial value

According to the latest data released by the National Bureau of Statistics, driven by technological innovation and commercial application, the scale of the artificial intelligence industry continues to grow. It is expected that by the end of the 14th Five-Year Plan, China’s AI-related industry scale will exceed 10 trillion yuan, moving towards a broader growth space.

According to statistics from relevant research institutions, the global artificial intelligence market size is expected to reach $757.58 billion by 2025, a year-on-year increase of 18.7%. As the world’s second-largest economy, China’s AI industry scale is exhibiting explosive growth driven by the synergy of policy support, market demand, and technological iteration.

At the same time, some foreign scholars have recently predicted the bursting of the AI bubble, believing that “if the AI boom collapses, the impact will be less than that of the internet bubble, but the shock will be widespread.” So, has a bubble really emerged in the AI field? In an interview with a reporter from “Daily Economic News,” Chen Lan stated that although AI company valuations are generally high at present, the practical application of AI technology has indeed improved efficiency and reduced costs. For example, Deloitte uses AI Agents to triple audit efficiency, significantly enhancing data processing and risk identification efficiency, and reducing the labor costs of a large number of repetitive tasks.

Chen Lan believes that bubbles typically occur when capital investment far exceeds the ability to realize technology and when revenue expectations are long disconnected. However, the ongoing creation of substantial value through AI applications means that it cannot simply be deemed to have a bubble at this time.

The “Report” pointed out that the development of AI in Asia presents structural characteristics of “coexisting tiers, differentiated paths, and great cooperation potential.” In the leading representative category, China has formed a mature full-chain capability and large-scale implementation capacity, while Japan and South Korea focus on high-end manufacturing and industrial automation; Singapore, as an application demonstration type, plays a governance demonstration and platform hub role; emerging economies like India and Indonesia, as potential cultivation types, leverage market potential and application scenarios as breakthroughs.

AI technology applications have shifted from point-driven to systematic integration

Huatai Securities’ research report pointed out that under the conflict between the US and Iran, global oil prices have surged, driving up overseas coal prices. Based on expectations, the price center of 5500 kcal thermal coal in northern ports of China will rise to around 750 yuan/ton by 2026.

According to calculations by Huatai Securities, if the price of 5500 kcal thermal coal at the port rises by 50 yuan/ton (including tax), the wholesale electricity price on the supply side will increase by 2.9%, corresponding to an industrial electricity price increase of 2.0% to 2.2%.

Will the rise in global electricity prices hinder AI development? Chen Lan told a reporter from “Daily Economic News” that the global shift of AI development focus from Europe and America to Asia is mainly due to factors such as talent, industrial policy, and application demand, rather than the power supply itself. While strained power supply does pose challenges to large-scale computing power deployment, stable power supply is particularly important for AI infrastructure.

“Currently, many countries facing energy shortages are also restarting nuclear power generation or investing in new energy to ensure the stable operation of computing power and data centers. However, AI development relies more on computing power optimization, data governance, and industrial policy rather than a single issue of power supply,” said Chen Lan.

In addition, focusing on the development of artificial intelligence in China, this year’s government work report has proposed for the first time to “create a new form of intelligent economy.” From “artificial intelligence +” to “creating a new form of intelligent economy,” what new signals are being released behind this?

In an interview with a reporter from “Daily Economic News,” Chen Lan stated that the transition from the “artificial intelligence +” proposed in previous years to the “intelligent economy” first proposed in this year’s government work report reflects an upgrade in the policy layer’s positioning of AI. “Artificial intelligence +” emphasizes the empowerment of AI technology in a single industry or business scenario, while “intelligent economy” is a systematic concept that emphasizes driving industrial upgrades, business innovation, and social service optimization with AI at its core.

“This releases an important signal that artificial intelligence is moving from the stage of technological breakthroughs to the stage of large-scale application, and the application of technology has shifted from point-driven to systematic integration. In the future, there will be more policies and resources supporting the construction of the intelligent economy, rather than just the promotion of AI technology itself. This also means that artificial intelligence will be more deeply integrated into enterprise production, operation, and service systems,” Chen Lan said.

Cover image source: Event organizer