Artificial intelligence has transformed the conversation around metals. After copper, uranium, and rare earth elements, silver is back in investors’ conversations. The logic seems compelling: more AI means more data centers, which means more semiconductors, more power grids, more solar panels, and more electronic equipment. Silver is one of the best-known electrical conductors. It is used in contacts, solder, conductive pastes, certain electronic components, and photovoltaic cells.

But the key question isn’t simply whether AI uses silver. It does, directly or indirectly. The real question is: how much silver can we actually attribute to AI’s needs based on the available data? This is where the analysis must remain rigorous. Part of the industrial demand for silver is well documented. Another part stems from a market narrative that is still difficult to isolate. For an individual investor, this distinction is essential. It is best to avoid confusing structural trends with narrative hype surrounding the price of silver in 2026.

Key Takeaways from the Article

• AI contributes to the demand for physical silver, but data centers are just one of the driving forces behind it.
• The data in current studies primarily measure silver usage by sector—electronics, electrical equipment, and solar—and do not adequately isolate the portion attributable to AI.
• Industrial demand for silver reached a record high in 2024, driven by several sectors, including certain AI-related applications (The Silver Institute).
• Global electricity consumption by data centers could nearly double between 2025 and 2030, requiring more electrical infrastructure (IEA)
• Public data does not currently allow for a precise quantification of the amount of silver used per AI server. Therefore, presenting a definitive figure is purely speculative.

Where is the silver in the AI ecosystem?

A data center is not a silver ingot in disguise. In a data center, silver is present in certain electronic components, contacts, and solder joints. Its conductive properties contribute to the operation of IT and electrical equipment. It is part of an infrastructure that includes servers, circuit boards, power supply systems, and backup devices. However, its presence does not allow us to determine a standard amount per facility: this depends on the equipment used and its design.

The connection to silver also involves electricity generation. When an operator contributes to the development of new photovoltaic installations to power its servers, it can support the demand for the metal used in solar cells. This effect remains indirect. Purchasing solar electricity does not necessarily mean installing solar panels. Furthermore, manufacturers are working to reduce the amount of silver required per cell. The growth of data centers can therefore support the use of this metal without automatically leading to a proportional increase in the volumes consumed.

How can we measure the amount of silver in a data center?

Available data allows us to track demand for silver by major sector, but not to precisely isolate the quantities destined for data centers. According to The Silver Institute, total global demand reached 1.16 billion ounces in 2024, across all uses. Industrial demand, meanwhile, reached a record high, driven in particular by electronics, electrical equipment, photovoltaics, and the automotive industry. Certain AI-related applications contribute to this trend, though these figures do not allow us to measure their exact share.

AI is therefore part of an already well-established industrial demand. Circuit boards, sensors, connectors, and relays were using silver long before its rise. The development of AI infrastructure may increase the need for equipment, but not all increases in silver consumption in these sectors can be attributed to it.

The solar industry illustrates another nuance: producing more equipment does not necessarily mean consuming more silver. According to market data for 2026, demand for silver from solar cell manufacturers is expected to decline due to a reduction in the amount of metal used in each cell. The growth of solar power—to which the electricity needs of data centers may contribute—does not, therefore, automatically translate into a proportional increase in the demand for silver.

Data Center Electricity Consumption

According to the International Energy Agency (IEA), data center electricity consumption could nearly double, rising from 485 TWh in 2025 to 950 TWh in 2030. This increase requires upgrades to power grids, power supplies, cooling systems, and backup systems.

The demand for silver is therefore not limited to chips. It also extends to the components and electrical contacts of the equipment that ensures the continuous operation of data centers.

However, this growth does not allow for a direct calculation of the quantities of silver required. Public data does not provide a reliable ratio between a data center’s power consumption and the amount of metal used.

AI: The Gap Between Actual Demand and Market Expectations

The link between AI and precious metals is capturing the markets’ attention: more data centers suggest a rise in demand for silver. But portraying this demand as massive and precisely measurable goes beyond what the available data shows.

AI supports several sectors that consume silver, notably electronics, electrical equipment, and photovoltaics. These sectors also serve the needs of other industries; therefore, their growth cannot be attributed entirely to the development of artificial intelligence. AI contributes to industrial demand, but it is impossible to isolate its specific share or precisely quantify the amount of silver used by each data center.

What role will AI play in the price of silver in 2026?

In 2026, the price of silver will depend on several factors. Electronics, solar energy, the automotive industry, and electrical grids are driving industrial demand. The development of AI may reinforce these needs by stimulating investment in data centers and their equipment. But this alone is not enough to predict price trends.

When silver becomes more expensive, manufacturers seek to reduce the quantities used, replace the metal whenever possible, or increase recycling efforts. An increase in equipment production therefore does not necessarily translate into a corresponding increase in silver consumption.

This significant role in industry distinguishes silver from gold. Its price reacts to interest rates, the dollar, and investor demand, as well as to global economic activity. AI thus represents one potential factor among others, the impact of which depends on actual demand and efforts to conserve the metal.

Conclusion

Artificial intelligence is driving demand for silver through electronics, electrical equipment, and solar technology associated with data centers. While these applications are real, the exact portion attributable to AI remains difficult to measure.

The Silver Institute documents the strength of industrial demand, while the IEA highlights the growing electricity needs of data centers. These trends suggest support for silver consumption, though they do not allow for quantifying the extent of this support. For an investor, AI is therefore a demand factor to monitor, though it does not guarantee a rise in the price. The analysis must also take into account available supply, recycling, metal savings, and the economic environment, all of which influence prices.