Key highlights
- 1The active electrical cables market is expected to reach USD 11 billion by 2036, at a CAGR of 25.6% from 2026 to 2036, driven by AI back-end networking and the migration to higher data rates, from a fast-growing base.
- 2AECs enable AI back-end networks. AECs deliver reliable, lossless interconnect for the remote direct memory access (RDMA) back-end networks that AI clusters are built on, with zero soft link flaps and higher signal integrity than passive copper.
- 3The only copper option for 3 to 7 metres. AECs maintain signal integrity at 56G and 112G per lane over about 0.5 to 7 metres, and are the only copper option covering the 3 to 7 metre range where passive direct-attach copper falls short.
- 4Copper beats optics on power and cost. Passive copper cannot reach far enough at high speed and active optical cables draw more power and cost more, so AECs fill the gap, and passive copper bundles are about 5 times heavier and nearly 3 times thicker than an AEC for the same links.
- 5Rates are migrating to 1.6T. AECs support 100G, 200G, 400G, 800G and 1.6T, and by early 2026 the category leader demonstrated live 1.6T AEC traffic and PCIe Gen6 retimer extensions.
- 6Demand is surging. The category leader reported about 272% year-on-year revenue growth in a recent quarter, driven by hyperscaler AEC demand.
- 7Key companies include Credo Technology Group, Amphenol Corporation, Molex, TE Connectivity, and Marvell Technology.
Report Overview
The active electrical cables market covers AECs, copper cables with retimer and signal-conditioning silicon in the connectors, used for high-speed interconnect in AI and data center networks, spanning data rates from 100G to 1.6T, scale-up in-rack and scale-out back-end connectivity, and hyperscale, cloud and enterprise applications. Passive direct-attach copper, active optical cables and optical transceivers are outside the scope except for comparison. AECs are the retimed-copper interconnect that bridges the gap between passive copper and optics in AI networks. Demand is shaped by AI back-end networking, lane-rate reach limits, power and cost, and reliability. This report examines the size, drivers, data rates, connectivity, applications, pricing, regions, competition, recent developments, and outlook of the market, and provides recommendations. Sizing is built bottom-up from AEC units and prices by data rate, connectivity and region, and reflects AECs deployed in AI and data center networks.
Market dynamics
Drivers
- 01AI back-end networking is the primary driver as AI clusters are built on lossless, low-latency remote direct memory access networks that connect GPUs and switches, and AECs provide reliable interconnect with zero soft link flaps, which is essential to these networks.
- 02The reach limits of passive copper are a technical driver as passive direct-attach copper cannot maintain signal integrity at higher lane rates over the distances AI racks require, and AECs, with retimers at both ends, extend copper to about 7 metres.
- 03The power and cost advantage over optics is a driver as active optical cables draw more power and cost more, and copper AECs fill the 3 to 7 metre gap at lower power and cost, which matters at cluster scale.
- 04Reliability is a driver as AECs deliver the lossless links AI training requires.
Opportunities
- 01Scale-up in-rack connectivity is an opportunity as GPU-to-GPU and GPU-to-switch links within and between racks use AECs where copper reach allows.
- 02PCIe and CXL retimer extensions are an opportunity as retimed copper extends high-speed PCIe and memory interconnect, with PCIe Gen6 retimer AECs demonstrated.
- 03Hyperscaler design wins are an opportunity in themselves as AEC demand is concentrated in hyperscalers building AI clusters.
Trends
- 01Rate migration is the defining trend as AECs move from 400G and 800G toward 1.6T, with 224G per lane on the horizon.
- 02Retimer silicon advances are a trend with new digital signal processing and smart retimer devices for 800G and 1.6T AECs, including the leader's Bluebird architecture and new entrants' retimer systems-on-chip.
- 03The displacement of DAC and AOC is a structural trend as AECs take the 3 to 7 metre range from heavier, shorter-reach passive copper and higher-power optics.
- 04Hyperscaler-driven demand is a trend reflected in the leader's rapid revenue growth on hyperscaler AEC orders.
Report Summary
| Base Year | 2025 |
|---|---|
| Forecast Period | 2026-2036 |
| Market Size (2025) | USD 900 million |
| Market Size (2026) | USD 1.13 billion |
| Market Size (2036) | USD 11 billion |
| CAGR (Value) | 25.6% (2026-2036) |
| Format | PDF & Excel |
| Segments Covered | By Data Rate: 400G & Below, 800G, 1.6T & Above. By Connectivity: Scale-Up (In-Rack), Scale-Out (Back-End). By End User; By Region. |
| Geographies Covered | North America, Asia-Pacific, Europe, and Rest of World |
| Key Companies | Credo Technology Group, Amphenol Corporation, Molex, TE Connectivity, Marvell Technology, Astera Labs, Broadcom Inc., Point2 Technology, Spectra7 Microsystems, Other Companies |
Segmental analysis
By Data Rate
- By data rate the market comprises 400G and below, 800G, and 1.6T and above. 800G holds the largest share, at about 55% of the market in 2026, with the remaining share divided across 400G and below and 1.6T and above. 800G is the largest rate because it is the volume rate for current AI back-end networks, where AECs are the emerging and increasingly dominant interconnect, running at 112G per lane over eight lanes. 400G and below serve earlier and lower-rate deployments and have an installed base. 1.6T and above is the fastest-growing band, as networks migrate to 1.6T and toward 224G per lane, with live 1.6T AEC traffic demonstrated by early 2026.
- The position of 800G reflects the current AI network build while 1.6T grows as the leading edge advances.
By Connectivity
- Scale-out back-end connectivity holds the larger share at about 60% of the market in 2026, with scale-up in-rack connectivity accounting for the remainder.
- Scale-out back-end connectivity linking racks to leaf and spine switches over the 3 to 7 metre range, is the larger segment because it is the sweet spot where AECs uniquely fill the gap between passive copper and optics in the RDMA back-end network.
- Scale-up in-rack connectivity linking GPUs and switches within and between adjacent racks over shorter reaches, is a large and growing segment as AI systems scale up.
The larger share of scale-out reflects the reach range where AECs are uniquely suited, while scale-up grows with dense AI system architectures.
By End User
- Hyperscale and AI hold the largest share at about 70% of the market in 2026, with the remaining share divided across cloud and enterprise.
- Hyperscale and AI are the leading end user because AEC demand is driven by hyperscalers building large AI clusters with lossless back-end networks, and demand is concentrated among them.
- Cloud providers adopt AECs for high-speed networking.
- Enterprise adopts AECs for AI and high-performance computing as those workloads spread.
The dominance of hyperscale and AI reflects that AEC demand is driven above all by large-scale AI cluster construction.
Geographic analysis
North America Active Electrical Cables Market
North America is the largest regional market driven by the concentration of hyperscale AI cluster construction and by leading AEC and retimer-silicon companies. The region leads in AI back-end network deployment and in the adoption of 800G and 1.6T AECs, and hosts the category leader and major hyperscalers. Strong AI investment and network scale make North America the leading and fast-scaling market for AECs.
Asia-Pacific Active Electrical Cables Market
Asia-Pacific is a major and fast-growing market with large AI data center construction in China and elsewhere and a strong cable, connector and electronics manufacturing base. The region supplies AEC assemblies and adopts them in AI networks, and hosts manufacturing for global suppliers. Rising AI capacity and the region's manufacturing strength make Asia-Pacific a significant and fast-growing market.
Europe and Rest of World
Europe is a growing market with AI and cloud data center construction adopting AECs for high-speed networking. The rest of the world, including the Middle East, adds growing demand as AI data center investment spreads. These regions add growth as AI networks are built globally, served by the leading AEC and retimer suppliers.
Pricing Analysis
Pricing in active electrical cables reflects data rate, reach, and silicon content. AECs are priced between passive direct-attach copper, which is inexpensive but limited in reach, and active optical cables, which cost and consume more, filling the 3 to 7 metre gap at attractive cost and power. Higher data rates and longer reaches, which require more capable retimer silicon, command higher prices, and the retimer, gearbox and forward-error-correction silicon is the main cost and value driver. Several factors set price. Data rate is central, as 1.6T AECs cost more than 800G and 400G for their more capable silicon. Reach affects price, as longer cables require stronger signal conditioning. Silicon content, the retimer and signal-conditioning devices, is the main cost and differentiator. Volume and hyperscaler contracts affect price, as large orders drive scale. Position against alternatives sets the ceiling, as AECs must remain cheaper and lower-power than optics while outperforming passive copper.
The trajectory of pricing depends on rate migration, silicon advances and scale, and the shift to higher rates raises value while volume and competition moderate price.
Competitive landscape
The market is led by the AEC pioneer and served by cable and connector companies and retimer-silicon providers. Credo Technology Group invented active electrical cables and leads the market with its HiWire and ZeroFlap AECs supporting 100G to 1.6T, its Bluebird digital signal processing architecture, and rapid revenue growth on hyperscaler demand. Amphenol, Molex and TE Connectivity, major cable and connector companies, supply AEC assemblies and interconnect. Retimer-silicon providers Marvell Technology, Astera Labs and Broadcom supply the digital signal processing and retimer devices at the heart of AECs, and Point2 Technology and Spectra7 Microsystems provide smart retimer systems-on-chip for 800G and 1.6T AECs.
Competition turns on retimer-silicon performance and power, signal integrity and reliability, data-rate leadership, and hyperscaler design wins, and the market combines the pioneer, cable and connector majors, and retimer-silicon specialists. AI back-end networking, rate migration and reliability favour suppliers with leading retimer silicon and proven hyperscaler deployments, and Credo holds a strong lead while connector majors and silicon providers compete and partner. The field is fast-growing and competitive, with rapid advances to 1.6T and 224G per lane as AI networks scale, and design wins with hyperscalers are decisive.
Voice of Customer
For our AI back-end network, AECs give us the lossless copper links we need in the three-to-seven-metre range where passive copper cannot reach and optics cost too much power. Zero link flaps matter enormously for training, and we are moving to 800G now and planning 1.6T, so retimer-silicon performance and reliability drive our choices.
Network architect, hyperscale operator (North America):
AECs are lighter and thinner than passive copper bundles for the same links, which helps airflow and cabling in dense racks, and they use less power than optics. We evaluate on data rate, reach and reliability, and we value suppliers with proven hyperscaler deployments and a roadmap to higher rates.
Interconnect engineer, cloud provider (Asia-Pacific):
AEC demand has surged with our AI cluster buildout, and supply and design wins matter. We work with the leading AEC and retimer suppliers, and the criteria are signal integrity, power, and a clear path to 1.6T and beyond as our networks scale.
Procurement lead, AI infrastructure builder (North America):
Analyst perspective
The active electrical cables market is one of the fastest-growing AI interconnect categories, created by the specific demands of AI back-end networks. The technical logic is clear: AI clusters are built on lossless, low-latency RDMA networks, and at higher lane rates passive copper cannot reach far enough while optics draw too much power and cost, so AECs, retimed copper cables that maintain signal integrity over 3 to 7 metres, fill the gap and are becoming the dominant interconnect in that range. The category leader invented AECs and has grown revenue rapidly on hyperscaler demand, and the market is migrating from 800G to 1.6T, with 224G per lane ahead. Cable and connector majors and retimer-silicon providers round out a fast-scaling ecosystem.
The honest considerations are the AI capital cycle, competition from optics, and concentration. The market's very high growth rests on continued heavy AI network buildout, which is cyclical, so a slowdown in AI capital spending would slow AEC demand, and the figures should be read as tied to the AI buildout. Competition comes from both sides: passive copper at the shortest reaches and optics, including linear-drive and co-packaged optics, at longer reaches and as power efficiency improves, so AECs must keep their power, cost and reach advantage in the middle range. Demand is concentrated among a few hyperscalers and design wins are decisive, so the market is exposed to customer concentration and to the pace of each rate transition. The market should be assessed on AI network buildout, rate migration and the copper-versus-optics balance rather than on the AI theme alone, and AI back-end networking and rate migration support very strong growth, with the AI cycle and competition from optics the key variables.
Key Strategic Developments
- 2024-2026: The category leader introduced 800G HiWire ZeroFlap AECs for AI back-end networks, delivering zero soft link flaps and higher signal integrity for lossless RDMA networks.
- November 2024: Point2 Technology announced a smart retimer mixed-signal system-on-chip for 800G and 1.6T AEC applications in hyperscale AI data centers.
- Early 2026: The category leader demonstrated live 1.6T AEC traffic and PCIe Gen6 retimer extensions, validating its next-generation digital signal processing architecture.
- 2025-2026: The category leader reported about 272% year-on-year revenue growth in a recent quarter, driven by hyperscaler AEC demand, reflecting rapid category adoption.
- 2024-2026: AECs increasingly displaced passive direct-attach copper and active optical cables in the 3 to 7 metre range of AI networks, emerging as a distinct and increasingly dominant interconnect category.
Strategic Recommendations
For AEC and retimer suppliers
The priority is to lead on retimer-silicon performance and rate migration and to win hyperscaler designs, because the market is driven by AI back-end networking and decided by silicon performance and design wins. Companies should advance to 1.6T and 224G per lane, improve signal integrity, power and reliability, extend retimed copper to PCIe and CXL, and secure hyperscaler and cloud deployments. Building supply and partnerships across cable, connector and silicon strengthen the position.
For data center operators
The recommendation is to adopt AECs for the 3 to 7 metre range of AI back-end networks where they offer the best balance of reach, power, cost and reliability, matching data rate to network roadmaps, and to weigh AECs against passive copper and optics by reach. For investors, this is one of the fastest-growing AI interconnect categories, to evaluate on AI network buildout, rate migration and the copper-versus-optics balance rather than on the AI theme alone, recognising that AI back-end networking and the migration to higher rates support very strong growth while the AI capital cycle, competition from optics, and customer concentration are the key variables.
Sustainability impact
Power Efficiency versus Optics
AECs consume less power than active optical cables over their reach range, reducing interconnect energy in AI networks.AECs support power efficiency.
By using retimed copper rather than optics over the 3 to 7 metre range, AECs consume less power per link than active optical cables, reducing the energy used by the vast number of interconnects in AI clusters and improving the efficiency of AI networks.
Material and Airflow Efficiency
AECs are lighter and thinner than passive copper bundles for the same links, improving airflow and reducing material. AECs support material and cooling efficiency.
By replacing heavier, thicker passive copper bundles, AECs reduce cabling material and improve airflow in dense racks, supporting cooling efficiency and reducing the material footprint of interconnect.
Enabling Efficient AI Compute
AECs enable the reliable, high-speed networks that let AI compute run efficiently at scale. AECs support efficient computing.
By providing lossless, high-speed interconnect, AECs enable AI clusters to run efficiently, supporting the networks that underpin AI while helping manage the power and reliability of large-scale compute.
Electronics Life Cycle and Recyclability
The sustainability of AECs depends on the life cycle of their copper and electronics and on recyclability. AECs require life-cycle stewardship.
By combining copper cable with retimer electronics, AECs depend on responsible sourcing and end-of-life recycling of copper and electronic components, so life-cycle stewardship is part of their sustainability in fast-refreshing AI infrastructure.
Table of contents
14 chapters · 278 pages · click to expandFrequently asked questions
The active electrical cables market was valued at USD 900 million in 2025 and is projected to reach USD 11 billion by 2036, growing from USD 1.13 billion in 2026, at a CAGR of 25.6% from 2026 to 2036, driven by AI back-end networking and the migration to higher data rates, from a fast-growing base.