Poesis, the Menlo Park AI-native asset manager founded by former Capital Group partner Alex Popa and former Goldman Sachs global head of machine learning Charles Elkan, said on 10 November 2026 that it plans to open Poesis AI Clarity, its flagship agentic US large-cap equity strategy, to external capital in early 2027. The release gives no performance figures, assets under management or fees. Fintechbits found no earlier coverage of Poesis in its archive. A Fintechbits search of SEC EDGAR full-text filings on 10 November 2026 found no Form D or Form ADV that mentions Poesis.

How Poesis AI Clarity works

According to the Poesis release, AI Clarity aims to capture equity market gains while seeking to materially reduce losses in downturns, an approach Poesis calls Dual Mandate Equity. AI agents carry out research, validate data and propose trades using quantitative models, and a dynamic beta overlay cuts market exposure in stressed markets and raises it in stronger ones.

Humans stay in charge. The chief investment officer and team set the strategy, risk framework and limits, and every trade proposal needs human review and approval before execution. Poesis describes the model as machine intelligence under human command.

In the autumn, Poesis also ran a one-week live experiment with ASUS and NVIDIA in which autonomous agents researched, managed risk and traded within set mandates, on a separate system with separate capital.

Popa spent more than 15 years at Capital Group, where he was a portfolio manager on The Growth Fund of America. Elkan led machine learning at Goldman Sachs and across e-commerce and cloud at Amazon, where he was the first Amazon Fellow, and is a longtime professor at UC San Diego.

Popa said: “AI now lets us apply that discipline across the full investment cycle with greater consistency, while keeping judgment and accountability with people.”

What investors will need to see

The release has no returns, no drawdown figures, no asset total and no fee schedule. For an institutional strategy, those are the first things an allocator asks for. A strategy designed to cut losses in downturns needs to show it did so through at least one real sell-off.

The human sign-off on every trade is a sensible control, but it also limits how fast and how often the strategy can trade. That suggests a lower-turnover, fundamentals-led approach more than a high-frequency one.

Fintechbits analysis

Why Poesis AI Clarity Will Be Judged on the Track Record

Poesis AI Clarity has a strong team and a fashionable story, and neither substitutes for performance. Institutional investors have seen many quantitative and machine learning funds promise downside protection, and most were judged on whether they delivered it when markets fell. The founders’ records at Capital Group and Goldman Sachs will get meetings; returns will win mandates.

The governance design is the most credible part. Keeping a person accountable for every trade addresses regulators’ concerns about autonomous AI in investment decisions and fits fiduciary duties. It also makes AI Clarity easier to explain to investment committees than a black box.

Fintechbits has covered AI moving into advice, including Evergreen’s AI financial advice push. Asset management is a harder test, because the results are public and measurable. An adviser chatbot can be judged on client satisfaction; an equity strategy is judged against an index every month.

Capacity is another question. Large-cap US equities are liquid, so the strategy could in principle take a lot of money, but a beta overlay that cuts exposure in stressed markets can lag sharp recoveries. Allocators will want to see both sides of that trade-off before committing.

Frequently asked questions

What is Poesis AI Clarity?

AI Clarity is Poesis’s agentic institutional equity strategy in US large-cap stocks, in which AI agents research and propose trades and humans approve each trade. Poesis plans to open it to external capital in early 2027.

Who founded Poesis?

Poesis was founded by Alex Popa, a former Capital Group partner and portfolio manager, and Charles Elkan, former global head of machine learning at Goldman Sachs, who is chief scientist.

Has Poesis published AI Clarity returns?

No. The 10 November 2026 release does not include performance figures, assets under management or fees.

What to Watch Next

The detail to watch is the first disclosure of Poesis AI Clarity performance, ideally including how it behaved during the autumn 2026 sell-off in large AI and chip stocks. That record will decide whether institutions commit capital when the strategy opens in early 2027. Fintechbits has tracked AI agents moving into wealth management elsewhere, from Focal AI’s adviser agents to bank-wide rollouts.