The Aura AI Photonics ETF (NYSEARCA:PHOX) arrives at a moment when investors want more granular ways to own the AI buildout than another mega-cap semiconductor basket. Optical interconnects, silicon photonics, and laser-based transceivers are becoming the bottleneck technology inside AI data centers, and PHOX is one of the first US-listed funds built specifically around that niche. For anyone who already owns the Magnificent 7 through an index fund, PHOX is being pitched as the missing puzzle piece: focused exposure to the companies moving light instead of electrons through the AI stack.
The question is whether a narrow, brand-new thematic ETF actually earns a slot in a portfolio, or whether a broader semiconductor fund already gives you the same exposure at a lower cost.
The Fund and the Problem It Solves
PHOX is a new ETF from the Tidal Trust III series, with Tidal Investments as the adviser and a prospectus dated July 29, 2026. It lists on NYSE Arca and carries total annual operating expenses of 0.60%. That fee sits above a plain vanilla semiconductor index fund but in line with most single-theme AI ETFs currently on the market.
The strategic problem PHOX is trying to solve is real. As hyperscalers push data center capacity toward gigawatt scale, moving data between GPUs, switches, and racks using copper is becoming a physics problem. Photonics, the use of light and optical components to transmit data, is how the industry is responding. Companies making optical transceivers, laser diodes, silicon photonics chips, and the networking gear that ties them together are positioned as picks-and-shovels suppliers to the AI infrastructure trade. Broad AI ETFs and even semiconductor indexes tend to underweight these names because their market caps are small relative to NVIDIA, Broadcom, or TSMC.
The return engine, then, is capital appreciation from a concentrated basket of photonics and optical networking suppliers whose revenue is tied directly to AI data center capex cycles. This is a growth-oriented, equity-only strategy. Income is not the point.
Does It Deliver?
PHOX is too new to evaluate on results. The fund only recently began trading and does not yet have a meaningful NAV history, distribution record, or tracking data investors can stress-test. Any article claiming otherwise is inventing numbers.
What can be evaluated is the fund’s structural promise against the alternatives. Broad semiconductor ETFs like the VanEck Semiconductor ETF or the iShares Semiconductor ETF give investors heavy exposure to NVIDIA, TSMC, Broadcom, and AMD, but photonics pure-plays typically account for only a low single-digit weight in those funds. If you believe the optical interconnect thesis is where the incremental capex dollar is heading, a diluted position inside a mega-cap semi ETF is not going to move your portfolio the way a concentrated basket would.
That is PHOX’s real argument. Whether it delivers depends on two things the prospectus alone cannot answer: whether the index or selection methodology actually captures the right companies (including the less obvious component suppliers, laser makers, and silicon photonics foundries alongside the well-known optical networking names), and whether the 0.60% expense ratio is worth paying versus buying three or four photonics stocks directly. For investors who do not want single-stock risk in a volatile subsector, the fund structure itself has value. For investors comfortable picking names, the fee is a real drag over time.
The Tradeoffs
Three constraints are worth naming before adding PHOX to a portfolio.
- Concentration and volatility. Photonics is a narrow subsector. Individual holdings can swing 20% or more on a single earnings report because these companies often derive a large share of revenue from a handful of hyperscaler customers. Expect drawdowns deeper than a broad tech index during any AI capex pause.
- No operating history. The fund launched in 2026 with no track record to evaluate tracking error, liquidity in stressed markets, or how the methodology performs across a full cycle. Bid-ask spreads on new thematic ETFs are often wider than the expense ratio itself in the first year of trading.
- Fee versus alternatives. A 0.60% expense ratio is reasonable for a specialty theme fund but meaningfully higher than the 0.35% or lower charged by mainstream semiconductor ETFs. Over a decade, that spread compounds against you if the photonics tilt does not add commensurate return.
There is also a thematic risk investors sometimes gloss over. If copper interconnects, co-packaged optics standards, or an entirely different networking architecture wins out faster or slower than the market expects, a photonics-focused basket can lead or lag the broader AI trade by wide margins in either direction.
The Close
PHOX makes sense as a small satellite position, roughly 2% to 5% of an equity sleeve, for investors who already own diversified AI or semiconductor exposure and specifically want to overweight the optical interconnect layer of the data center buildout. It is a targeted satellite bet best sized accordingly. Investors looking for one-fund AI exposure are better served by a broader semiconductor or AI infrastructure ETF, where photonics is present but not the whole story. And anyone who cannot stomach 30%-plus drawdowns in a concentrated subsector should look elsewhere entirely, because the same narrow focus that makes PHOX interesting on the way up is what will make it painful to hold during any pullback in AI capex sentiment.
The fund’s real test will come with its first full year of returns and its first significant market drawdown. Until then, PHOX is a thesis wrapped in a ticker, and the thesis is only as good as an investor’s conviction in photonics as the enabling layer of the next decade of AI infrastructure spending.
Contact [email protected] for any questions or corrections.
Top Gaining Stocks
Top Losing Stocks