Over five years, buying AI stocks directly probably beats RVI for most retail investors — because a ~2.2% fee plus a 10–30% NAV premium quietly drains 4–6 points of return a year. That's the short answer. The longer answer is more interesting.
RVI — the Robinhood Ventures Fund — is one of the only ways a retail brokerage account can touch late-stage private names like OpenAI, SpaceX, and Stripe before they IPO. That access is real, and it's rare. But access isn't the same as return. The question isn't "can I own private AI?" It's "after fees, premium, and tax, do I end up with more money than if I'd just bought Nvidia and Microsoft?" Let's run it.
RVI vs AI Stocks: The 5-Year Return Comparison
RVI vs buying AI stocks directly comes down to a fee-and-access trade. RVI gives you diversified pre-IPO exposure for roughly a 2.2% expense ratio plus whatever premium to NAV the market is charging — often 10–30%. A direct AI basket gives you liquid, transparent public names for near 0% in fees but no private upside. Over five years, the fee-and-premium drag usually outweighs the private kicker unless RVI's book delivers a marquee exit.
| Attribute | RVI (Robinhood Ventures) | Direct AI Stocks |
|---|---|---|
| Annual fee | ~2.2% expense ratio | ~0% (commission-free trades) |
| Entry cost vs value | 10–30% premium to NAV | Market price = fair value |
| What you own | Pre-IPO private companies | Public AI leaders (NVDA, MSFT, AVGO) |
| Liquidity | Limited; can trade at a discount | Instant, deep daily volume |
| Transparency | Quarterly NAV marks, lagged | Real-time pricing + filings |
| Diversification | ~20–40 private positions | As broad as you build it |
| Tax treatment | Fund-level gains + distributions | You control timing of gains |
| Upside driver | Exits / IPOs re-rating NAV | Earnings + multiple expansion |
The Fee and Premium Math Nobody Runs
Start with a $10,000 investment in each. The direct AI basket costs you essentially nothing to hold — a Nvidia, Microsoft, Broadcom, and Alphabet split runs $0 in commissions and you pay tax only when you sell. RVI is different. A 2.2% expense ratio is about $220 in year one, and because it compounds on a (hopefully) growing base, it totals roughly 10–11% of your capital over five years even if the fund just matches the market.
Then layer the premium. If you buy RVI at a 20% premium to NAV, you've paid $12,000 of market price for $10,000 of underlying assets. That premium may persist — or it may collapse to a discount, as closed-end and listed venture vehicles often do once the novelty fades. Historically, many closed-end funds drift to 5–15% discounts. If RVI's premium swings from +20% to −10% over five years, that's a 30-point headwind on top of fees, independent of how the portfolio performs.
~2.2%
RVI annual expense ratio
10–30%
Typical NAV premium at entry
~0%
Cost to hold direct AI stocks
When RVI Actually Wins the 5-Year Race
RVI is not a bad product — it's a leveraged bet on private exits. The fund wins the five-year race in a specific scenario: one or more of its anchor holdings IPOs or gets acquired at a large step-up, dragging NAV up faster than public AI multiples expand. If OpenAI lists at 2–3x its last private mark, or SpaceX prints a long-rumored $400B+ IPO, RVI's NAV could jump in a way no single public stock would replicate, because you bought that exposure at the private valuation.
Public AI stocks, by contrast, are already priced for the boom. Nvidia trades on consensus AI demand; the easy re-rating from $500B to $3T+ in market cap has largely happened. Their next five years depend on sustaining 20–40% earnings growth against rising expectations — real, but not the same asymmetry as a pre-IPO mark converting to a public price. That's the genuine case for RVI: you're paying the fee and premium to rent access to step-ups you can't otherwise buy.
The catch is concentration and timing. If exits stall — and the IPO window has been narrow since 2022 — RVI's NAV stays marked at stale private valuations while you keep paying 2.2% a year. You can track which names are actually nearing the public markets on the Tech IPO tracker and gauge how private AI is being priced on the AI Valuations dashboard.
RVI vs AI Stocks: Three Realistic 5-Year Scenarios
Base case — AI keeps compounding
Public AI returns ~12–15%/yr; RVI matches NAV growth but loses ~4–6 pts/yr to fees + premium decay.
Direct AI stocks win by a clear margin.
Bull case — big private exits
Two or three anchor holdings IPO at 2–3x marks; RVI NAV jumps 40–60% in a single window.
RVI wins — the private kicker overwhelms the fee drag.
Bear case — IPO window stays shut
Marks go stale, premium flips to a discount, fees keep compounding at 2.2%/yr.
Direct AI stocks win decisively on liquidity and cost.
Two of three realistic scenarios favor buying AI stocks directly. RVI only wins when the private-exit machine fires — which is exactly the outcome that's been hardest to count on for the past four years.
The Verdict: Which Should You Own?
Buy AI Stocks Directly If…
- ✓ You want liquidity and near-zero fees
- ✓ You value real-time pricing and transparency
- ✓ You want to control the timing of your gains for tax
- ✓ This is core, not satellite, capital
Consider RVI If…
- ✓ You specifically want pre-IPO exposure
- ✓ You can buy near NAV, not at a 20–30% premium
- ✓ You're sizing it as a small satellite position
- ✓ You can stomach illiquidity and stale marks
For most retail investors over five years, the math points one way.
Buy the AI stocks directly. Treat RVI as a small, opportunistic bet on private exits — bought near NAV, never at a 30% premium.
Compare private and public AI exposure on the AI Valuations and Tech IPO dashboards at Value Add VC. Originally published in the Trace Cohen newsletter.