---
title: "💢💢💢"
type: "Topics"
locale: "en"
url: "https://longbridge.com/en/topics/40446090.md"
description: "📌 $AMD(AMD.US) + GlobalFoundries pulling $SIVE into the core ecosystem? This "laser source position" might not be priced in by the market yet. The market is currently discussing CPO, with most people watching: $NVIDIA(NVDA.US)TSMBroadcomMarvell. But if we look one layer further down to the "even more fundamental" level, a variable that's easier to overlook starts to appear: the light source. And $SIVE happens to be positioned right there. The key isn't how big it is, but—where it appears. First layer of logic: being written into the ecosystem map..."
datetime: "2026-05-05T23:45:42.000Z"
locales:
  - [en](https://longbridge.com/en/topics/40446090.md)
  - [zh-CN](https://longbridge.com/zh-CN/topics/40446090.md)
  - [zh-HK](https://longbridge.com/zh-HK/topics/40446090.md)
author: "[辰逸](https://longbridge.com/en/profiles/16318663.md)"
---

# 💢💢💢

📌 $AMD(AMD.US) + GlobalFoundries pulling $SIVE into the core ecosystem? This "laser source position" might not be priced in by the market yet.

The market is currently talking about CPO, with most people focusing on:

$NVIDIA(NVDA.US)  
$Taiwan Semiconductor(TSM.US)  
Broadcom  
Marvell

But if we look one layer deeper into the "more fundamental level," a variable that's easier to overlook starts to appear:

The light source.

And $SIVE happens to be positioned right here.

The key isn't how big it is, but—where it appears.

First layer of logic: Being written into the ecosystem map, earlier than signing contracts

GlobalFoundries, in its ecosystem presentation, lists Sivers Semiconductors as "one of two public laser suppliers."

This in itself is already not an ordinary signal.

Because this kind of "ecosystem map" is essentially:

A preview of the future roadmap  
A list of scalable partners  
The framework for a potential mass production system

It doesn't represent revenue already generated, but it represents "being selected into the system."

And in the semiconductor industry, this step often occurs:

Before orders  
Before scale  
Even before market recognition

Second layer of logic: The real bottleneck for CPO is not the GPU

The core change of CPO isn't just about placing the optical module closer to the chip.

The real challenge lies in:

How to make the light source stable, low-power, and scalable  
How to make it compatible with packaging processes  
How to integrate it at the system level

This turns the "laser source" from a component into a system-level critical node.

Not a simple replacement.

In other words:

Without a stable light source, CPO is just a design blueprint.

Third layer of logic: The $AMD(AMD.US) + GFS path is bypassing the traditional route

The focus of AMD's collaboration with GlobalFoundries isn't advanced process nodes, but:

Specialized processes  
Packaging and photonic-electronic integration  
CPO-related exploration

This path is different from the $NVIDIA(NVDA.US) + $Taiwan Semiconductor(TSM.US) route.

And on this path:

GFS isn't just a foundry, but a "platform integrator."

This means it will define the ecosystem.

And once the ecosystem is defined:

The number of light source suppliers won't be large  
Companies that make the list will see their weight amplified

This is the significance of $SIVE appearing on that map.

Fourth layer of logic: Why hasn't the market reacted yet?

Because there are three "natural blind spots" here:

No clear order disclosure  
No breakdown of revenue by segment  
CPO is still in its early stages

So it's hard for institutional models to be built:

No orders → can't forecast revenue  
No revenue → can't assign a valuation

Thus, it gets ignored.

This is similar to the logic mentioned earlier:

Not the main character  
But positioned at a critical interface

Fifth layer of logic: What does a $1.7 billion market cap mean?

If $SIVE's market cap is around ~$1.7 billion,

And the CPO light source is a "must-have component" with a limited number of suppliers,

Then if just one thing happens:

Entering the actual mass production chain

The valuation elasticity could be very high.

But conversely:

If it remains stuck in the "ecosystem map stage,"

Then it's just a "possibility asset."

Sixth layer of logic: What to watch isn't the present, but the validation nodes

The most crucial thing here isn't judging "will it benefit,"

But watching for a few validation signals:

Whether it enters specific projects (even if the customer isn't named)  
Whether CPO-related revenue descriptions appear  
Whether it's repeatedly mentioned by more ecosystem partners  
Whether it enters the hyperscaler (cloud vendor) chain

Once these signals appear:

The market will go from "not seeing it" to "repricing."

And this switch often happens very quickly.

The current question isn't:

Whether $SIVE is a core player,

But:

When the $AMD(AMD.US) + GFS CPO route truly materializes,

Will the light source position be highly concentrated,

Or will it be quickly replaced?

Do you lean more towards this being a "key node pre-positioned early," or a "seemingly critical but ultimately marginalized link"?

### Related Stocks

- [GFS.US](https://longbridge.com/en/quote/GFS.US.md)
- [TSM.US](https://longbridge.com/en/quote/TSM.US.md)
- [AVGO.US](https://longbridge.com/en/quote/AVGO.US.md)
- [AMD.US](https://longbridge.com/en/quote/AMD.US.md)
- [AMDL.US](https://longbridge.com/en/quote/AMDL.US.md)