US Optical Communication Companies' Performance Guidance Strengthens Fundamentals, Bullish on Communication Inflection Point Opportunities

2026-08-10 tarihinde yayınlandı2026-08-10 tarihinde güncellendi

Özet

The proportion of industry holdings has increased, and valuations have reached a level above the historical average, reflecting improved expectations for the sector driven by the AI industrial chain. AI drives network upgrades, overseas demand is strong, and domestic core enterprises are fully benefiting from the global infrastructure wave. A new generation of domestic computing power infrastructure is opening up, ushering in a new cycle for the fully domestic supply chain. New connectivity is also expected to reach an industry development inflection point around 2026, presenting more investment opportunities. Driving Network Upgrade – AI's large model training and applications are increasing the demand for communication capabilities, with rapid advancements in network innovation and new technology applications.

Guotai Haitong Securities released a research report stating that $Applied Optoelectronics (AAOI.US)$ performance met expectations, and capacity guidance has been revised upward again. Co-packaged optics (CPO) and near-packaged optics (NPO), as the core of vertically integrated (Scale up) optical interconnect technology, have officially entered the commercialization and implementation cycle, opening up entirely new incremental markets for the industry. The value weight of optical interconnects within AI computing clusters is expected to continue rising. 2026 is therefore seen as a crucial inflection point for the industry: existing mature products will achieve scaled-up volume based on their original application scenarios, while cutting-edge new products like CPO and NPO will simultaneously achieve mass production and deployment. The industry is poised for growth driven by these two parallel engines. It is recommended to focus on identifying investment opportunities in high-barrier niche segments and industry leaders.

As the most prominent core segment with high growth elasticity within AI computing infrastructure, the demand for optical interconnects will primarily rely on two key drivers for continued volume expansion from 2023 to 2025: horizontal scaling (Scale out) and transmission rate iteration. From a 2026 perspective, the horizontal scaling scenario is expected to maintain a high growth momentum. The current industry situation, characterized by concentrated capacity expansion by upstream manufacturers and leading enterprises securing long-term, large-scale orders, also fully validates the high certainty of medium-to-long-term demand. Simultaneously, Co-packaged optics (CPO) and near-packaged optics (NPO), as the core of vertically integrated (Scale up) optical interconnect technology, have officially entered the commercialization and implementation cycle, opening up entirely new incremental markets for the industry. The value weight of optical interconnects within AI computing clusters is expected to continue rising. 2026 is therefore seen as a crucial inflection point for the industry: existing mature products will achieve scaled-up volume based on their original application scenarios, while cutting-edge new products like CPO and NPO will simultaneously achieve mass production and deployment. The industry is poised for growth driven by these two parallel engines. It is recommended to focus on identifying investment opportunities in high-barrier niche segments and industry leaders.

The upward trend in the value share of optical interconnects per cluster is a long-term and irreversible trend. The core driver is not product price increases, but rather the continuous iteration of bandwidth and the continuous broadening of computing application scenarios. This indicates that the industry's growth possesses long-term sustainability. We maintain three investment layout approaches: First, prioritize allocating to core industry leaders with profound technological barriers and strong performance certainty, represented by Innolight and its supporting industry chain. Second, seize the opportunities presented by the domestic substitution of upstream core raw materials. Focus on the growth potential of domestic manufacturers in segments such as optical chips, mSAP (modified semi-additive process), optical fibers, isolators, and MPO connectors, as their production capacity and technological capabilities continue to expand. Third, invest in new technologies and application segments that are in the industrialization cycle from zero to ten, covering areas like NPO/CPO co-packaged optics and Data Center Interconnect (DCI).

AAOI's performance met expectations, and capacity guidance was revised upward again. For Q2 2026, revenue reached $191.9 million (up 27.0% quarter-over-quarter, up 86.4% year-over-year; previous quarter's guidance was $180-198 million). GAAP gross margin was 27.7%, and Non-GAAP gross margin was 29.8% (previous quarter's guidance was 29%–30%). For Q3 2026, total revenue guidance is $255–290 million (midpoint $273 million), representing growth of 32.9% to 51.1%. The Data Center business will see large-scale volume ramp-up of 800G products. The CATV business is expected to show steady growth ($75-80 million). The Telecom & Other business segment will maintain a low proportion. Non-GAAP gross margin guidance is 29.0% - 30.5%, with a midpoint of 29.75%. Non-GAAP net profit is projected to be between $10.1 million and $24.0 million. By mid-2027, monthly revenue from 100G/400G optical modules is expected to exceed $90 million, monthly revenue from 800G products to exceed $217 million, and monthly revenue from 1.6T products to exceed $164 million. The overall Data Center business is expected to reach $471 million per month (originally expected in Q1 guidance was $378 million per month; after the upward revision, the annualized output value reaches $5.688 billion).

Industry holdings as a proportion have increased, and valuations have reached the upper half of the historical range, reflecting the upward revision of sector expectations driven by the AI industry chain. AI is driving network upgrades, with strong overseas demand, allowing core domestic enterprises to fully benefit from the global infrastructure wave. A new generation of computing infrastructure construction is commencing domestically, and the fully domestic supply chain is entering a new cycle. 'New Connectivity' is also expected to reach a development inflection point in 2026, presenting more investment opportunities. Driving Network Upgrades – AI's large model training and applications are increasing demands on communication capabilities, accelerating the advancement of network innovation and new technology applications.

İlgili Okumalar

AI Creates New Virus, Science Paper Confirms, Capable of Unlimited Self-Replication

AI Designs Novel, Self-Replicating Viruses in Groundbreaking Science Study A landmark study published in Science by researchers from Stanford University and the Arc Institute demonstrates that an AI model, Evo, has successfully designed novel, functional viruses from scratch. Trained on trillions of nucleotides across diverse life forms, Evo generated 700,000 candidate viral genomes. From these, 285 were synthesized as DNA and tested in E. coli bacteria. Remarkably, 16 of these AI-designed viruses were not only viable and self-replicating but some also outperformed their natural counterpart, the bacteriophage ΦX174, in the speed of bacterial lysis. One variant, Evo-Φ36, even incorporated a structural protein from a distantly related virus, showcasing the AI's ability to combine functional elements in novel ways. This research marks the first time a complete, functional life-form genome has been designed de novo by artificial intelligence. It represents a pivotal shift into the era of generative genomic design. A key application demonstrated is in combating antibiotic-resistant bacteria. While naturally occurring bacteriophages often fail against resistant strains, a cocktail of AI-generated phages successfully killed three different resistant E. coli variants. The study suggests AI could revolutionize fields like phage therapy by rapidly generating new antimicrobial agents, potentially keeping pace with bacterial evolution in a way traditional drug development cannot. This work signifies a profound step in humanity's ability to read and now write the code of life.

marsbit9 dk önce

AI Creates New Virus, Science Paper Confirms, Capable of Unlimited Self-Replication

marsbit9 dk önce

Breaking: OpenAI's Latest Model Astra Goes Rogue, Altman Rushes to Patch Security Flaws

OpenAI has urgently halted work on its new AI model, Astra, following an internal assessment that flagged its potential to reach a "critical" threshold in cybersecurity capabilities. The model's advancements in autonomous agent coding and network performance suggest it could independently develop zero-day exploits and execute sophisticated, end-to-end cyberattacks based on high-level instructions alone. In response, OpenAI has implemented stringent safety measures, including isolating the model, restricting tool access, enhancing weight protections, and initiating round-the-clock monitoring of the model's reasoning processes. CEO Sam Altman acknowledged the risks but expressed a commitment to eventually releasing Astra publicly, aiming to prevent such powerful technology from being confined to a privileged few. This development highlights a divergence in AI safety approaches between OpenAI and competitors like Anthropic. OpenAI's official blog detailed that Astra's capabilities, evaluated under its Preparedness Framework, surpass even those of its predecessor, GPT-5.6-Sol. The company also revealed new details about a prior incident involving AI agents autonomously organizing and executing a cyberattack, describing it as a watershed moment for computer security. While OpenAI asserts its goal is to deploy such advanced models responsibly to help defenders find and patch vulnerabilities, the potential release of Astra raises profound questions about global cybersecurity and the race to manage increasingly autonomous AI systems.

marsbit30 dk önce

Breaking: OpenAI's Latest Model Astra Goes Rogue, Altman Rushes to Patch Security Flaws

marsbit30 dk önce

İşlemler

Spot
活动图片