The Copper Limit Everyone Is Building Around Is About to Break

The Copper Limit Everyone Is Building Around Is About to Break

by Salah Nasri (host), with guest Amit Shaligram

Applied Sciences & Engineering Economics & Finance KNIndustry and industrial studiesKNATExtractive industries
🎙 Salah Nasri (host), with guest Amit Shaligram 👥 5K 📅 July 8, 2026 ⏱ 53 min 👁 276 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

silicon photonicscopper limitAI data centersoptical transceiverssemiconductor scaling

Summary

In this podcast episode, host Salah Nasri interviews Amit Shaligram, Director of Marketing and Applications Engineering at STMicroelectronics. They discuss the imminent limits of copper interconnects in AI data centers, predicting a massive shift to silicon photonics as optical transceivers scale from millions to hundreds of millions of units. Shaligram explains ST’s differentiated approach, including 300mm wafer manufacturing and a low-loss edge coupling solution. He highlights the challenges of standardization in the industry, the importance of ecosystem development, and the potential for silicon photonics to become foundational across semiconductor technologies. The conversation also covers Shaligram’s career journey, from co-founding a startup to working at a large corporation, and his insights on managing multi-billion dollar sales funnels. He touches on the convergence of technologies across AI, LEO satellites, and automotive, and argues that the feared AI power ceiling may not materialize due to disruptive innovations. The episode concludes with reflections on leadership and the importance of people in the semiconductor industry.

160 words

Critical Evaluation

Value of the Information & Strength of the Argument

The value of the information lies in the guest’s insider perspective on the semiconductor industry, particularly the shift from copper to silicon photonics in AI data centers. Shaligram provides concrete examples of ST’s technological differentiators, such as the low-loss edge coupling solution and 300mm wafer manufacturing. The argumentation is coherent and grounded in his professional experience, though it lacks quantitative data or references to specific studies. The discussion on the power ceiling and disruptive technologies is thought-provoking but remains at a high level without deep technical analysis.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is moderate; the conversation is based on expert opinion rather than peer-reviewed research. No specific sources are cited within the video, and the description does not provide links to external references. The title accurately reflects the content, focusing on the copper limit and its impending break. The discussion is consistent with industry trends, but the lack of citations limits its scientific robustness. No comments were provided for analysis.

173 words

Title / Content Match

The title accurately reflects the central theme: the limitations of copper in AI data centers and the shift to silicon photonics.

Quality & Reliability

7/10

The discussion is based on the guest's professional experience and industry knowledge, but lacks detailed technical data or citations. Claims about silicon photonics scaling and power ceilings are plausible but not backed by specific studies or sources.

Chapters

Contribution & Novelties

The episode provides an industry insider’s perspective on the transition from copper to silicon photonics in AI data centers, highlighting STMicroelectronics’ specific technological advantages and the challenges of standardization. It offers insights into the scale of optical transceiver demand and the potential for silicon photonics to become a foundational technology.

Pour aller plus loin :

98 words

Radar Profile

The radar profile shows a balanced performance across all dimensions, with slightly higher scores in quantity and quality of information, and lower in technical depth. This indicates a well-rounded discussion that is informative but not overly technical, suitable for a broad audience interested in semiconductor trends.

Reliability 7/10