September 2026 mARKet Update Summary

Moderated by Portfolio Manager Dan White, CFA, ARK's September Market Update webinar opened with a macroeconomic overview by Cathie Wood, ARK's Chief Executive Officer and Chief Investment Officer. The broader discussion then turned to signals to watch for the next yield curve inversion, new agentic AI products and inference-speed economics, Tesla's Cybercab launch, new additions to the ARK Venture Fund, the urban air mobility opportunity, and the first positive Phase 3 readout for a personalized cancer vaccine. View the full webinar [HERE]

If we have another inversion, how does an investor distinguish a deflationary-boom inversion from a recessionary one in real time?

Cathie explained that strong real gross domestic product (GDP) growth, driven by AI, should dominate the picture even as another yield curve inversion approaches. She pointed to Anthropic's annualized revenue run rate accelerating from ~$9 billion in December to ~$65 billion in July, and to reaccelerating growth among hyperscalers, as evidence that this kind of surprise is likely to broaden across companies on the right side of change. At the same time, she expects deflationary busts in sectors being disrupted, such as transportation and telecommunications, and flagged private credit and private equity, which have financed mature, cash-generative companies with short-term floating-rate debt, as a potential weak link if rates continue rising. Cathie believes GDP statistics, still largely rooted in industrial-age measurement, will be revised upward as the new growth regime becomes more visible, even as debate continues over whether specific data points reflect boom or bust.

We often talk about model releases, but two recently launched AI products, GrokBot and Muse, represent a different kind of release. Could you explain what they are and how they differ?

Frank Downing, Director of Research, AI and Cloud, explained that GrokBot (from xAI/SpaceX) and Muse (from Meta) are agentic products rather than new underlying models, extending an architecture of persistent, proactive, easy-to-message agents that ARK first saw hints of earlier this year. Both products message users proactively rather than only responding to prompts, but GrokBot skews more business-focused while Muse is built for consumers, arriving with deep out-of-the-box integrations across email, calendar, and financial accounts. Frank noted that both offer free and paid subscription tiers similar to other AI products, and that Muse in particular marks a notable milestone for Meta, whose underlying models have reached a level of capability the company can now directly monetize through a consumer product.

Speed has turned into its own axis in AI. Labs are designing their own inference chips and specialists are pushing token throughput past what graphics processing units (GPUs) deliver. Why does speed matter this much now, and who captures the value?

Frank noted that inference speed has become a genuine point of competition as specialized chipmakers like Cerebras build systems that run models many times faster than standard GPU-based infrastructure. He cited Cerebras' CS3 system, which is running OpenAI's latest model roughly 14 times faster than GPU-based deployments, compressing a 20-minute agentic task, such as building a slide deck, into under two minutes and keeping users in a productive workflow. Beyond agentic tasks, he pointed to voice interaction, where faster inference avoids awkward pauses as models "think," and latency-sensitive use cases like quantitative trading, citing Jane Street as an early adopter willing to pay for the fastest available intelligence. Frank observed that OpenAI and Anthropic already charge roughly double for "fast mode" inference at double the speed, and he expects an even steeper premium for the ultra-fast tier now emerging.

Please share ARK's thoughts on the recent Cybercab event.

Tasha Keeney, CFA, Director of Research, Autonomous Technology & Robotics / Director of Investment Analysis, discussed Tesla's introduction of Cybercab to its robotaxi fleet, built using an "unboxed" manufacturing process that removes hydraulic braking fluid and simplifies the vehicle's design. She noted that Tesla also disclosed reaching ~1 million unsupervised miles, a meaningful acceleration from the figure given on the prior earnings call, which ARK is watching closely as a signal for the pace of Cybercab's manufacturing rollout. Tasha explained that Cybercab's lower operating cost per mile is central to ARK's thesis that robotaxis will expand the ride-hail market well beyond what today's human-driven fleets can support, giving Tesla an advantage over both traditional ride-hail and other autonomy providers running more expensive vehicles.

Please highlight a new company in the ARK Venture Fund.

Brett Winton, Chief Futurist, ARK Venture Investment Committee Member, highlighted K2 Space, a vertically integrated satellite manufacturer building the high-power satellites needed to run AI workloads in orbit and to serve customers, including the Department of Defense. He also pointed to Cosm, which builds immersive, dome-based venues for live sports and entertainment that ARK believes will enhance rather than replace the value of in-person events, and Wave, which licenses autonomous vehicle software to multiple manufacturers and has gained early traction in Europe. Brett added that Boring Company, also part of the portfolio, is pursuing partially evacuated tunnels between US cities that could enable travel speeds of ~700 to 800 miles per hour, an ambition that Cathie noted has also come up in discussions of lunar infrastructure as Elon Musk's focus expands toward the moon.

Archer is close to commercial launch and Dubai is already mapping out air taxi routes. What is ARK's long-term view on urban air mobility?

Tasha explained that ARK expects autonomous electric vertical take-off and landing (eVTOL) air taxis to become roughly price-competitive with today's ground taxis at scale while cutting travel time meaningfully, though widespread robotaxi adoption could offset some of that advantage as ground traffic evolves. She noted that defense applications, rather than urban air mobility itself, have become the more significant near-term opportunity for companies like Archer, which is developing a loyal-wingman aircraft and a logistics aircraft for the US military, sizes that defense opportunity at ~$100 billion, and recently acquired Boeing's Insitu business for ~$200 million as Boeing took a stake in the company. Tasha believes the defense opportunity meaningfully de-risks Archer's path in a heavily regulated industry while the broader urban air mobility market continues to scale.

If Tesla Semi cause supply chain disruption, will its economic value exceed Robotaxi?

Tasha estimated the global robotaxi opportunity at $10 trillion or more, with the broader autonomous trucking and logistics opportunity roughly double that, at ~$20 trillion, suggesting that Tesla Semi's total addressable market could ultimately exceed Robotaxi's. She noted that Tesla has signaled plans to introduce Semi autonomy by the end of this year or next, and that an autonomous, electric Semi could reduce logistics costs to roughly one-third of today's human-driven diesel trucking cost per ton-mile, a shift she believes the market needs, given an existing truck driver shortage.

The first individualized cancer vaccine just cleared Phase 3. What makes it different from everything that came before it, and is this the moment multiomics and AI stop being a thesis and start becoming standard of care?

Ovid Amadi, PhD, Multiomics Portfolio Manager and Director of Research, explained that the Moderna/Merck INTerpath trial combines two previously separate immuno-oncology approaches, checkpoint inhibitors such as Keytruda and CAR-T-style T-cell engineering, by sequencing a patient's tumor to identify ~34 patient-specific neoantigens, encoding them into an mRNA therapy that trains the patient's own T cells to attack the tumor, and pairing that therapy with Keytruda so that those T cells are not shut down. The combination outperformed Keytruda alone on both recurrence-free survival and distant metastasis-free survival in a trial of more than 1,100 high-risk melanoma patients. Ovid noted that the result required a convergence of technologies ARK has long tracked: less costly DNA and RNA sequencing, RNA manufacturing, and AI-driven selection of the most promising neoantigen targets.

Ovid also noted that Tempus was selected as Moderna's tumor-sequencing partner, applying its existing therapy-selection sequencing capability directly to neoantigen identification for vaccine manufacturing, while Personalis served as Moderna's sequencing partner throughout the clinical trials. He added that Twist Bioscience's oligonucleotide manufacturing and 10x Genomics' single-cell sequencing technology also support the broader ecosystem behind personalized cancer vaccines. Cathie added that Tempus' pending acquisition of Personalis further validates the platform, extending Tempus' reach into longitudinal sequencing data that spans well beyond oncology.

Conclusion:

September's discussion reinforced a theme running across ARK's innovation platforms: capabilities that lived as research theses are converting into measurable, monetizable products. AI progress is no longer just about model releases, it now shows up in agentic products and in the economics of inference speed. Autonomous vehicles are moving from software demonstrations toward manufacturing and unit-cost curves, with Cybercab and Tesla Semi illustrating how cost per mile, not just capability, will determine the pace of disruption. And in multiomics, the first positive Phase 3 readout for a personalized cancer vaccine shows how converging advances in sequencing, RNA manufacturing, and AI can compress a research thesis into a standard of care. At ARK, we remain focused on identifying the companies enabling these platform shifts before their economic magnitude is fully appreciated by the market.

Make sure to check out our In The Know video series for deeper insights.

Important Information

Investors should carefully consider the investment objectives and risks as well as charges and expenses of an ARK Fund before investing. This and other information are contained in the ARK Funds’ prospectuses and summary prospectuses, which may be obtained by clicking here. The prospectus and summary prospectus should be read carefully before investing. 

 An investment in an ARK Fund is subject to risks and you can lose money on your investment in an ARK Fund. There can be no assurance that the ARK Funds will achieve their investment objectives. The ARK Funds’ portfolios are more volatile than broad market averages. The ARK Funds also have specific risks, which are described in their respective prospectuses. 

 The information provided in this material is for informational purposes only and should not be used as the basis for any investment decision and is subject to change without notice. It does not constitute, either explicitly or implicitly, any provision of services or products by ARK, and investors should determine for themselves whether a particular investment management service is suitable for their investment needs. All statements made regarding companies or securities are strictly beliefs and points of view held by ARK and are not endorsements by ARK of any company or security or recommendations by ARK to buy, sell or hold any security. Historical results are not indications of future results. 

Certain of the statements contained in this material may be statements of future expectations and other forward-looking statements that are based on ARK's current views and assumptions and involve known and unknown risks and uncertainties that could cause actual results, performance or events to differ materially from those expressed or implied in such statements. ARK assumes no obligation to update any forward-looking information. ARK and its clients as well as its related persons may (but do not necessarily) have financial interests in securities or issuers that are discussed. Certain information was obtained from sources that ARK believes to be reliable; however, ARK does not guarantee the accuracy or completeness of any information obtained from any third party.

Yield Curve is a graphical representation of the interest rates on debt for a range of maturities. It shows the yield an investor is expecting to earn if they lend their money for a given period of time. An inverted curve appears when long-term yields fall below short-term yields. An inverted yield curve occurs due to the perception of long-term investors that interest rates will decline in the future.

“Hyperscalers” are cloud providers, such as AWS, Microsoft Azure, or Google Cloud that operate massive, globally distributed data centers designed for extreme scalability and elastic, on‑demand computing capacity.

 To view the top ten holdings for ARKK click here.

 To view the top ten holdings for ARKQ click here.

 To view the top ten holdings for ARKW click here.

 To view the top ten holdings for ARKG click here.

 To view the top ten holdings for ARKF click here.

 To view the top ten holdings for ARKX click here.

 To view the top ten holdings for ARKVX click here.

 Holdings are subject to change

 Foreside Fund Services, LLC, distributor.