Credo Technology Adds DustPhotonics, Building an AI Optical Connectivity Stack
1h ago|4 min readStandard
Fazen Markets Editorial Desk
Collective editorial team · methodology
CRDOoptical connectivity
Fazen Markets Editorial Desk
Collective editorial team · methodology
Trades XAUUSD 24/5 on autopilot. Verified Myfxbook performance. Free forever.
Risk warning: CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. The majority of retail investor accounts lose money when trading CFDs. Vortex HFT is informational software — not investment advice. Past performance does not guarantee future results.
Credo Technology Group Holding Ltd (CRDO) is developing a deeper portfolio for AI data center connectivity by acquiring and integrating the core technology of DustPhotonics. The move, announced in mid-June 2026, aims to create a comprehensive optical connectivity stack capable of addressing critical bandwidth bottlenecks in hyperscale and AI training clusters. Credo's strategy targets high-speed optical modules, a key component for linking AI servers, where market demand is projected to exceed $10 billion by 2028. The integration enhances Credo's ability to supply complete connectivity solutions, moving beyond its established role as a provider of high-speed SerDes chips and retimers.
The AI boom has created unprecedented demand for data transfer speed and energy efficiency inside data centers. The primary bottleneck for scaling AI training clusters has shifted from compute to the network fabric connecting thousands of GPUs. Optical connectivity, which uses light to transmit data, is the only technology capable of meeting the required bandwidth and distance needs for next-generation AI systems. This demand surge triggered a wave of industry consolidation; Credo's move follows Marvell's 2025 acquisition of Inphi for $10 billion, which similarly aimed to integrate optical component expertise with high-speed digital chips.
The current macro backdrop features intense competition among hyperscalers like Microsoft, Amazon, and Google to deploy larger, more efficient AI clusters. This arms race drives capital expenditure directly into networking hardware that can reduce AI model training times. Credo's integration of DustPhotonics technology is a direct response to an industry-wide push for higher-density, lower-power optical interconnects. The catalyst is the imminent transition to 1.6 Terabit per second (Tbps) optical modules, which require tight co-design of analog optical components with digital signal processors where Credo holds an edge.
Credo's stock (CRDO) traded near $28.50 following the announcement, giving the company a market capitalization of approximately $4.7 billion. The company reported fiscal year 2025 revenue of $211.3 million, a 19% year-over-year increase, with a gross margin of 62.4%. The broader market for high-speed optical modules used in AI is forecast to grow at a compound annual rate of 25%, reaching $10.8 billion by 2028 according to LightCounting research. For comparison, the Philadelphia Semiconductor Index (SOX) is up 14% year-to-date, while CRDO shares have gained over 40% in the same period, reflecting investor focus on AI-enabling infrastructure.
Credo's strategic shift is evident in its historical revenue mix and forward projections. Prior to this move, over 70% of Credo's revenue came from standalone connectivity chips like retimers and SerDes IP. The integrated optical module market commands significantly higher average selling prices, often exceeding $1,000 per unit for 800G and 1.6T products, compared to $50-$200 for a standalone retimer chip. This pivot aligns the company with direct competitors like Broadcom and Marvell in a higher-value segment. The total addressable market for Credo's combined offerings is now estimated at over $15 billion.
The integration positions Credo to compete directly with established optical module vendors like Coherent Corp. and II-VI (now Coherent) and vertically integrated giants like Broadcom. Credo's potential market share gain in the 800G/1.6T module segment could pressure gross margins at these peers. The move is also a positive for semiconductor capital equipment firms like Applied Materials and ASML, as the complex optical-electronic integration requires advanced packaging technologies like silicon photonics, driving demand for related tools.
The primary risk to Credo's strategy is execution complexity and customer qualification timelines, which can span 12-18 months in the conservative data center market. A counter-argument is that hyperscalers may prefer to design their own custom optical modules, a path pursued by Meta and Google, which could limit the total available market for merchant suppliers like Credo. Institutional positioning data shows increased option flow in CRDO calls, indicating a bullish bias among sophisticated investors anticipating design wins with major cloud providers in the second half of 2026.
Credo's fiscal Q1 2027 earnings report, expected in late August 2026, will provide the first commentary on integration progress and any early customer engagements for the new optical products. Key technical levels for CRDO stock include support near the 50-day moving average at $26.80 and resistance at the year-to-date high of $31.45. A break above $32 on high volume could signal institutional conviction in the strategy's commercial traction.
The next major industry catalyst is the Optical Fiber Communication Conference (OFC) in March 2027, where product demonstrations and performance benchmarks will be critical for competitive positioning. Investors should monitor inventory levels at key customers like NVIDIA and major hyperscalers; any slowdown in AI infrastructure spending would disproportionately affect suppliers in the optical connectivity chain. Credo's success hinges on securing at least one tier-1 cloud provider as a design win for its integrated modules before the end of 2026.
For retail investors, Credo's strategy represents a higher-risk, higher-potential-reward pivot. The company is moving from a component supplier model into a more complex systems market with larger deal sizes but fiercer competition. Success could lead to significant revenue acceleration and multiple expansion, while failure could result in costly R&D write-offs and lost market share in its core chip business. The stock's volatility is likely to increase around earnings announcements and news of customer wins.
Copper electrical cables face severe signal degradation over distances longer than a few meters at the multi-hundred-gigabit speeds required for AI. Optical connectivity uses lasers and fiber optics to transmit data as light, enabling much longer reach (hundreds of meters to kilometers) with far lower power consumption and signal loss. This is essential for connecting the sprawling, warehouse-scale AI clusters where thousands of servers must communicate simultaneously during model training.
The integration of optics and electronics on a single chip or module, known as silicon photonics, has been a long-term industry goal for over a decade. Intel has invested heavily in the technology since the early 2010s. The recent surge is driven by AI's specific demands, making the economics viable. Prior large-scale attempts, like Cisco's acquisition of Luxtera in 2019, were aimed at different markets, but the current AI-driven volume provides the scale needed for these technologies to become mainstream and cost-effective.
Vortex HFT is our free MT4/MT5 Expert Advisor. Verified Myfxbook performance. No subscription. No fees. Trades 24/5.
Position yourself for the macro moves discussed above
Start TradingSponsored
Open a demo account in 30 seconds. No deposit required.
CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.