Ideas
Innovation platforms drive historic investment cycle.
AI is accelerating all five ARK innovation platforms - robotics, energy storage, AI, autonomous mobility, and public blockchain - creating the largest investment cycle in history, larger than the late-1800s railroad buildout. This can lift global GDP growth to high single digits in the second half of the decade and make disruptive innovation the majority of equity market cap, while non-innovation exposures may be flat or down; investors should get exposure, including via ARK's actively managed ETFs.
AI compute drives trillion-dollar data center spend.
The world is starving for compute power. Foundation model providers like Anthropic and OpenAI are turning down business because they lack enough compute and data centers. Core AI alone can justify more than $1 trillion of spend by 2030, and data center systems spending could triple to $1.4 trillion; demand elasticity is stretching into the solar system.
AI power needs boost storage, solar, nuclear.
Energy supply is a key constraint and opportunity for AI data centers. Distributed energy is important; modular nuclear reactors should come on in the latter half of the decade; solar is delivering all marginal energy supply growth; and behind-the-meter AI compute needs batteries attached to balance power for chips.
SpaceX leads in space compute opportunity.
Terrestrial data center buildout faces execution challenges, so sending compute into space with SpaceX can be cost competitive or superior. SpaceX has a decade lead in reusable rockets, is the largest position in ARK's venture strategy, and could IPO this year as AI compute increases launch volume 10-60x, requiring many more rockets and launch towers.
AI chip demand supports multiple semiconductor winners.
Demand for AI chips is so extreme that there is room for multiple players despite Nvidia's dominance. Vertically integrated competitors include Tesla producing chips for SpaceX, Amazon, OpenAI with Broadcom, Google's TPU, and AMD. Nvidia is interesting, but its exceptional gross margins are unlikely to persist perpetually because chips remain cyclical.
Tesla robotaxi economics expand addressable market.
Robotaxis will be a remarkable scalable innovation on Tesla's platform and delivered robotaxi services could be worth $30 trillion collectively over the decade. Tesla's low-cost ~$35,000 cars and cheaper sensor approach versus Waymo should produce superior economics; at $1 per mile the addressable market increases more than 10x.
Chinese automakers excel in autonomous mobility scaling.
As autonomous mobility develops, natural geographic monopolies will likely form and even superior technology will be forced or incentivized to partner with local providers. Chinese manufacturers, because of their vehicle volume and ability to quickly scale new models and sensor sets, have a meaningful ability to deliver value in autonomous mobility.
AI lowers drug costs, lifting biotech.
AI will decrease the costs to deliver drugs to market and increase returns on R&D, benefiting drug discovery and biotechnology over the second half of the decade, although progress can be volatile with three steps forward and one step back.
This Bloomberg Markets video, published January 22, 2026,
features Brett Winton
discussing ARKK, ARK actively managed ETFs, DTCR, AI compute, Modular nuclear reactors, ICLN, SOLAR, SPCX, SMH, TSLA, Chinese autonomous vehicle manufacturers, XBI, AI-enabled drug discovery.
8 trade ideas extracted by AI with direction and confidence scoring.
Speakers:
Brett Winton
· Tickers:
ARKK,
ARK actively managed ETFs,
DTCR,
AI compute,
Modular nuclear reactors,
ICLN,
SOLAR,
SPCX,
SMH,
TSLA,
Chinese autonomous vehicle manufacturers,
XBI,
AI-enabled drug discovery