50x storage later, Justin Sun is always looking at the next decade

Most people in China first got to know Justin Sun through his eccentric stories.

He paid $30 million to have lunch with Warren Buffett, only to cancel at the last minute citing a kidney stone. He bought a banana taped to a wall for $6.2 million and ate it in front of everyone at the press conference. He invested $75 million to become the largest backer of the Trump family's crypto project, earning a seat at the White House dinner. At 35, he flew past the Kármán line and declared himself the youngest Chinese commercial astronaut.

He also has no shortage of negatives. In 2023, he was sued by the SEC for market manipulation, accused of over 600,000 wash trades to inflate TRX's price and hiring celebrities to promote without disclosing compensation. Currently, he and the Trump family's WLFI project are suing each other.

These stories are so widely circulated that they almost overshadow a serious fact: Over the past decade in the secondary capital market, this man has hardly missed betting on any major trend.

Starting to buy BTC in 2013, by 2016, he was advising post-90s not to buy houses, but to buy:

Bitcoin, Nvidia, Tesla, Tencent.

A decade later, as of May 2026, Tesla's total return is approximately 2,683%, and Nvidia's total return is nearly 24,000%.

If you had listened to Justin back then and invested 10,000 yuan in Nvidia, it would be worth 2.4 million today; 10,000 yuan in Tesla would be 278,000 yuan. An audience member who put 200,000 yuan into each item on that list in 2016 would see the Nvidia stake alone become about 48 million yuan, the Tesla stake about 5.4 million yuan, totaling over 53 million yuan.

And this man is still making moves today. On November 6, 2025, Justin Sun dropped a statement:

"Short-term chip shortage, long-term energy shortage, always storage shortage."

The capital market's reaction to this statement only turned fervent in 2026. SanDisk (SNDK), spun off from Western Digital, surged from a low of about $35 to $1,439 in a year, a maximum increase of nearly 50x.

HBM memory capacity from the three major manufacturers Samsung, SK Hynix, and Micron is fully booked, sold out for 2026, with orders extending into 2027-2028.

Just as everyone was still chasing the storage concept frenzy, in early 2026, Justin Sun shifted his stance again in a video.

That video was originally about his 2026 outlook. Besides health-related topics like putting wellness on the new year's list, he specifically spent time focusing on areas for young people: Embodied AI, Drones, Spatial Computing, and Space Exploration.

The author collected Justin Sun's public statements on these four directions over the past two years. Looking at them together, each path has seen the emergence of its initial capital leaders.

Who is the Next Storage Stock?

The first area Justin Sun pointed to was Embodied AI.

The concept of robots has been discussed by humans for at least a hundred years. Czech playwright Karel Čapek coined the word "Robot" in 1920. Industrial robotic arms have been used since the 1980s, and Honda's ASIMO could walk up and down stairs over twenty years ago. But the real bottleneck has always been the "brain."

In the last two years, the entire industry has shifted to VLA models: Vision-Language-Action. Simply put, robots that used to act based on code are now starting to perceive and act based on the world.

Unitree shipped over 5,500 humanoid robots in 2025, ranking first globally. In March 2026, it filed an IPO application on the STAR Market. Galaxy Bot secured a new funding round of $300 million (approx. 21 billion RMB) in December 2025, bringing cumulative funding to about $800 million and a valuation of $3 billion (approx. 211 billion RMB), setting dual records for the largest single round and cumulative funding in the Embodied AI track.

Justin Sun said he likely won't build robots himself, but he has a nose for this kind of narrative and capital flow. In an interview with Bloomberg, he once said, "In a market where 99% of people don't know what a wallet is, the cost of education must be factored into the business model."

This was true for stablecoins in 2018 and remains applicable in 2026. 99% of Chinese people haven't used an Embodied AI robot yet. But once that robot can cook, move boxes, or look after the elderly, the remaining 1% represents the next big opportunity.

The second track he pointed to was Drones.

While humanoid robots are still scaling up production, drones have already reached commercial maturity earlier. They are naturally suited for AI tasks—from autonomous navigation and swarm coordination to data collection, AI excels at it. Drones don't need to walk; flying is inherently simpler than navigating a humanoid form.

On the battlefield in Ukraine, AI drone swarms have already taken over a large part of the role of traditional tank units, with Ukraine's annual drone production target climbing into the millions. Over the rice paddies of rural China, DJI's agricultural drones are flying, each replacing ten farm workers. In Shenzhen, Meituan has already operationalized drone deliveries, with orders arriving in under 15 minutes.

Drones have advanced ahead of humanoid robots. They represent the first form of AI to achieve a closed commercial loop in the physical world.

The third area Justin Sun pointed to was Spatial Computing. This is the least mainstream of the directions he mentioned.

When Apple released the Vision Pro in 2024, most people dismissed it as an overpriced VR headset. This might be a misreading.

The Vision Pro's ambition has little to do with VR. It's Apple's first attempt to make AI understand space: the size of your living room, the distance of the table from you, whether the coffee cup is to the left or right of the sofa, whether you can reach it. This sounds simple but is ten times harder than training ChatGPT. Large language models only need to understand language; spatial computing needs to understand physics.

This is precisely the common prerequisite for robots, drones, and autonomous driving: they all require spatial intelligence. Nvidia's Cosmos platform, Google's Genie 3 world model, Tesla's FSD—they are all doing the same thing: transitioning AI from understanding text to understanding the world.

ChatGPT is smart enough understanding language, but the next generation of AI needs to understand the world itself.

While Justin Sun only talked about the first three tracks, he physically went for space.

On August 3, 2025, he sat inside Blue Origin's "New Shepard" capsule for mission NS-34 and flew past the Kármán line.

After returning to Earth, he expressed an ambition: hoping his company would no longer just be a "cryptocurrency exchange" but an "infrastructure service provider for the space economy," using blockchain to solve space asset rights verification, satellite data trading, and interplanetary payments. It sounds like science fiction. But look back at him evangelizing USDT ten years ago—people thought that was science fiction too.

Back on Earth, his message to young people was more direct, "Space exploration is the common mission of all humanity. I hope this flight can inspire more young people to engage in technology and innovation, jointly shaping humanity's interstellar future."

Justin Sun's Investment Logic

Justin Sun's publicly expressed investment logic is: find a track with a clear direction, lay out bets on both ends, and don't gamble on a single company's execution.

For the robotics track, his framework is to bet on the body and the brain separately.

Betting on the body with Tesla. In early 2026, Tesla announced the discontinuation of the Model S and Model X to convert the Fremont factory for Optimus production, targeting an annual output of one million units at a unit cost of around $20,000-$25,000. The current Optimus version is already handling parts handling and sorting at the Austin and Fremont factories, with the Gen 3 production line starting in summer 2026.

Betting on the brain with Nvidia. The Jetson Thor packs server-grade AI inference into the robot itself, and Isaac GR00T has become an industry-wide foundational platform. Jensen Huang declared at GTC a goal of 1 billion humanoid robots globally by 2035.

Whether Optimus delivers on schedule is Musk's problem, not Nvidia's. As long as the track runs successfully, the toll road will collect its fees.

For the drone track, the core judgment is the irreversibility of Physical AI in military scenarios.

AeroVironment's Switchblade loitering munitions became iconic weapons in Ukraine, with monthly production capacity ramping from 40 to 500 units, targeting 1,200 units, with a $3.9 billion order backlog securing revenue for the next three years. Kratos' XQ-58 Valkyrie serves as the "Loyal Wingman" for the F-35: manned aircraft handle the mission, drones run the flanks, at a fraction of the cost of a fifth-generation fighter. Kratos was up 280% in 2025 and another 72% in 2026.

One company makes tanks uneconomical; the other makes manned aircraft redundant. The logic of the two bets complements each other.

For the space track, Justin Sun won a Blue Origin flight seat for $28 million in 2021, donating the money to Blue Origin's STEM philanthropic fund distributed to 19 non-profits. On August 3, 2025, he completed a suborbital flight on the New Shepard mission NS-34.

In public markets, SpaceX submitted a confidential IPO draft to the SEC in April 2026, targeting a valuation of $1.75 trillion, which would be the largest IPO in human history. Rocket Lab's Q1 2026 revenue exceeded $200 million, making it the most direct substitute stock when SpaceX is unavailable.

Once SpaceX goes public, the pricing benchmark for the entire space sector will be rewritten.

Listen to Justin Sun's Words

Looking at Justin Sun's statements over the past two years together, "AI, robotics, and blockchain have reached their iPhone moment" is his judgment on Embodied AI. "Robot armies, robot police" is his prediction for autonomous weaponized AI. "The convergence of AI, robotics, and spatial computing" is his bet on the next generation of human-computer interfaces. "Earth is too small, it's our home" is his perspective shift after crossing the Kármán line.

Put together, these four pieces form the complete picture of Physical AI.

Over the past twenty years, the internet changed how information flows. WeChat replaced letter writing, Taobao replaced going to the market, TikTok replaced TV.

But the underlying rules of the physical world remained unchanged: workers were still workers, factories were still factories.

Over the next twenty years, AI might change the operating mode of the physical world itself. Factories will have tireless humanoid robots, roads will run with autonomous vehicles, battlefields will roar with drone swarms, and the first "residents" on the Moon and Mars will likely be advanced AI robots.

The young man who told everyone not to buy houses in 2016 has now flown past the Kármán line.

And most of us might still be waiting for the next Yanjiao.