Around 57,000 to 64,000 metric tons of gold sit in known underground reserves, ready to be mined with current technology and at current prices. That’s roughly a quarter of the 244,000 metric tons of gold that have been discovered on Earth so far. The rest, about 187,000 metric tons, has already been pulled from the ground over the course of human history.
Known Reserves vs. Total Resources
The distinction between reserves and resources matters here. Reserves are deposits that miners can extract profitably with today’s technology and gold prices. Resources include gold we know about (or strongly suspect exists) but can’t yet mine economically. The U.S. Geological Survey’s 2025 estimate puts proven global reserves at about 64,000 metric tons, up slightly from the 57,000-ton figure reported a few years earlier as rising gold prices made previously marginal deposits worth pursuing.
Beyond those reserves, there are an estimated 183,000 additional metric tons of gold in identified resources worldwide. These deposits are real, but extracting them would require higher gold prices, better technology, or both. The United States alone holds roughly 33,000 tons of gold resources, with nearly half of that in deposits that haven’t been formally identified yet. Globally, the total is far larger. So while “57,000 to 64,000 tons” is the headline number, the actual amount of gold left in the Earth’s crust is several times that.
How Long Current Reserves Will Last
Global gold mines produce about 3,000 to 3,600 metric tons per year. At that pace, the London Bullion Market Association estimates that current proven reserves would sustain production for roughly 14 years. That sounds alarming, but it’s a snapshot, not a countdown. Mining companies constantly convert resources into reserves as gold prices rise, extraction technology improves, and exploration uncovers new deposits.
When you include the broader resource base (those 183,000 additional tons), there’s enough gold to keep production going for another 46 years beyond that 14-year window. And exploration continues to find new deposits. The more realistic concern isn’t that gold will suddenly run out but that production will gradually flatten and decline. A lack of investment in exploration and the shift toward deeper, harder-to-reach underground projects points toward a flat or slowly declining production profile in the coming decades rather than a dramatic cliff.
Where the Remaining Gold Is
Gold reserves are concentrated in a handful of countries. As of 2025, the top five nations by unmined gold look like this:
- Russia: 12,000 metric tons
- Australia: 12,000 metric tons
- South Africa: 5,000 metric tons
- Indonesia: 3,800 metric tons
- Canada: 3,200 metric tons
Russia and Australia together hold nearly half of the world’s known reserves. South Africa, once the dominant gold producer on the planet, has seen its output decline for decades as its deepest mines become increasingly expensive and dangerous to operate. Indonesia’s reserves are largely tied to massive copper-gold deposits, while Canada’s are spread across numerous smaller operations. Geopolitics plays a real role here: sanctions, trade relationships, and domestic mining policies all affect how quickly these reserves actually reach the market.
Recycling Fills a Growing Share
New mining isn’t the only source of gold entering the market each year. In 2023, total gold supply reached about 4,900 metric tons. Of that, roughly 3,644 tons came from mines and 1,237 tons came from recycling, mostly old jewelry, electronics, and industrial scrap. That means recycled gold accounted for about 25% of the year’s total supply, and that share grew 9% from the previous year.
This recycling buffer is significant. Because gold is virtually indestructible and almost all of the 187,000 tons ever mined still exists in some form, the world has a massive above-ground stockpile to draw from. As mining becomes more difficult and expensive, recycling will likely pick up an even larger share of annual supply.
Gold Beyond Traditional Mining
The ocean floor holds gold in polymetallic nodules, potato-sized rocks scattered across the deep seabed that contain a mix of valuable metals. Mining companies have eyed these deposits for decades, but no commercial deep-sea mining operation has ever extracted them. The technical challenges are enormous, the environmental risks are poorly understood, and the primary metals of interest in those nodules are cobalt, nickel, and copper, not gold. Ocean gold isn’t going to meaningfully change the supply picture anytime soon.
There’s also gold dissolved in seawater at incredibly low concentrations, roughly 13 billionths of a gram per liter. The total amount is theoretically vast, but extracting it would cost far more than the gold is worth. For practical purposes, land-based mining and recycling will remain the only meaningful sources of gold for the foreseeable future.
Why “Running Out” Is the Wrong Frame
Gold mining won’t end with a sudden stop. It works more like squeezing a sponge: the easy gold came out first, and each additional squeeze yields less. The richest, most accessible deposits have largely been found. New discoveries tend to be deeper underground, in more remote locations, or at lower ore grades, meaning you need to move more rock to get the same amount of gold. The average grade of gold ore mined today is significantly lower than it was a few decades ago.
Rising gold prices can unlock deposits that were previously too expensive to mine, effectively increasing reserves without anyone discovering a single new ounce. This is why reserve estimates have actually grown over time even as mining continues. The 64,000-ton figure from the USGS in 2025 is higher than earlier estimates, not because geologists found huge new deposits, but because higher prices made marginal deposits economically viable. The real constraint going forward isn’t the amount of gold in the ground. It’s the cost, energy, and environmental impact of getting it out.