Announced capacity is not available capacity: normalizing InP and silicon photonics expansion claims before an RFQ

The decision

No. An announced capacity expansion is evidence of intent, not of an alternate source a buyer can quote today. Before an RFQ, normalize each candidate's public record into dated line items: the exact deliverable, material and wafer size, whether the capacity is current or planned, expected activation, acceptance scope, and access prerequisites. SMART Photonics' expected 2028 six-inch indium phosphide ramp and GlobalFoundries' plan to scale AMF's platform to 300mm are forward-looking statements on different material platforms, and neither can backfill a current InP process-step gap.

What each announcement actually commits

SMART Photonics states on its production page that it is a pure-play manufacturer offering the complete production process of photonic integrated circuits, from epitaxial growth and re-growth through processing and polishing, and that it supports customers from proof of concept to high volume production via three steps: design with a Process Design Kit, alpha production, and volume production. The same page states the company is investing in new six-inch indium phosphide wafer production capacity, and that when fully activated in 2028 the new production is expected to manufacture more than 12,000 wafers per year, corresponding to millions of photonic chips. Two qualifiers matter here: the figure is an expectation, not an operating result, and the page itself is undated, so a buyer cannot tell from the document alone when the statement was made or whether the schedule has since moved.

GlobalFoundries announced on November 18, 2025, with a Singapore dateline of November 17, 2025, that it had acquired Advanced Micro Foundry, a silicon photonics foundry. GlobalFoundries states the acquisition will expand its silicon photonics technology portfolio, production capacity and research and development in Singapore, and that it will address demands in long-haul optical communications, computing, LiDAR and sensing on AMF's 200mm platform in Singapore, with plans to scale to 300mm as market needs grow. GlobalFoundries also states the acquisition establishes it as the largest silicon photonics pure-play foundry by revenue. That ranking is the company's own assertion in its own announcement; the supplied sources do not independently verify it, and a public announcement is not independent proof of the underlying performance.

Why the two platforms are not interchangeable evidence

This is an inference from the two statements, not a conclusion either source draws. SMART Photonics' expected 2028 capacity is indium phosphide; GlobalFoundries' AMF platform is silicon photonics on 200mm wafers. A buyer whose constrained step is an InP process cannot treat a silicon photonics expansion, however credible, as evidence of an additional InP source: the material platforms, wafer sizes, process design kits and qualification paths differ. Likewise, GlobalFoundries states it plans to scale to 300mm as market needs grow, and the announcement excerpt does not quantify that demand, while SMART's more than 12,000 wafers per year is an expectation tied to a 2028 activation. Neither statement establishes capacity a buyer can purchase today, utilization, yield, qualification status, customer commitments, or whether either supplier would accept the reader's volume or process. A supply-gap plan that counts either announcement as present-day capacity would be built on forward-looking language.

The packaging and test boundary

Before comparing suppliers, a buyer should fix the process step being purchased, because the sources draw the boundary differently. PHIX states that it offers assembly services and contract manufacturing for photonic integrated circuits and MEMS, building optoelectronic modules on indium phosphide, silicon photonics, silicon nitride, and planar lightwave circuit platforms, and describes itself as the photonic equivalent of an OSAT company. PHIX also states that a photonic integrated circuit by itself is not yet a functional product: it needs connection to optical fibers, other PICs, cooling solutions, and electronics before it becomes part of a photonics-enabled module. SMART Photonics' scope, by contrast, is wafer-level manufacturing. A quoted wafer and a quoted assembled module are different deliverables with different acceptance criteria, and a requirement document should say which one the buyer needs and who accepts it.

Access prerequisites before an RFQ

AIM Photonics' Multi-Project Wafer and Test, Assembly and Packaging agreements page illustrates how much commercial groundwork precedes any reservation. AIM states that purchasing wafer fabrication or test, assembly and packaging services involves disclosure of confidential information, so the first step is an agreement covering that disclosure; members are already covered by their membership agreement, while non-members must have an executed non-disclosure agreement. AIM then requires a services agreement setting payment and delivery terms before it can reserve space on an MPW run or a private run or provide test, assembly and packaging services, plus an export control form to meet U.S. export control regulations. AIM also states its agreements are executed through the Research Foundation for the State University of New York on behalf of the SUNY Center for Economic Development. Treat this sequence as an access-prerequisite checklist, not as evidence of technical qualification: signing an NDA does not qualify a process.

A proposed capacity-evidence ladder

This ladder is an Atom Industries proposal, built from the four sources, for normalizing public evidence before an RFQ. Record each candidate in a dated table with these rows: source and its publication date or undated status; exact deliverable (epitaxy, wafer fabrication, packaging or assembly, or test); material and wafer size; process stage; capacity status marked plainly as current or planned; expected activation date; acceptance and test scope; and access prerequisites such as NDA, services agreement and export control forms. Ask each candidate to state whether the quoted step is epitaxy, fabrication, assembly or test, and who accepts the delivered item. Treat any announced expansion as a separate line item from current purchasable capacity, and require a dated update from the supplier before the expansion enters a supply-gap schedule. Where a source is undated, note that explicitly and ask for the statement's date in the first nonconfidential exchange.

Conditions that would change the approach

The approach above would change if a supplier provides dated, written confirmation that announced capacity is already producing, along with qualification evidence and volume by quarter showing that the buyer's requirement can fit. It would also change if the buyer's requirement shifts platform, for example from indium phosphide to silicon photonics, which would make GlobalFoundries' AMF platform relevant rather than adjacent. If the constrained step is assembly rather than wafer fabrication, PHIX's stated scope, chip-to-chip hybrid integration, fiber array coupling, and DC and RF interfacing, becomes the comparison set and the wafer announcements matter less.

Atom Industries' position and the next question

Atom Industries is development-stage, originating a potential manufacturing acquisition, joint venture or expansion that an experienced operator could deliver. It has no operating facility, qualified process, signed customer, committed financing, or available production capacity, and nothing in this article should be read as an Atom capability or as a relationship between Atom and any supplier named here. The next question a supply-chain lead should ask each candidate is simple and nonconfidential: of the capacity you have announced, what portion is producing today, on which platform and wafer size, and what is the earliest date you would accept a new requirement at a stated volume by quarter?

Sources & further reading

This perspective supports research and relationship development. It does not represent available manufacturing capacity, a committed transaction or an investment offering.