Americas semiconductor industry and key debates ahead of the Communacopia + Technology Conference 2026 Report Interpretation

Overview

This conference preview sets out Goldman Sachs’ expectations for management commentary across digital compute, semiconductor equipment, memory and storage, analog, and EDA software. The central conclusion is that AI-led demand should support a broad, though debated, semiconductor upcycle, with supply, capacity, competition and valuation transmission the key issues for investors to track.

Core views

For digital compute, Goldman Sachs expects companies to maintain a bullish tone on AI demand. Robust hyperscaler spending and agentic AI are expected to support infrastructure investment, while falling token costs and tangible returns in use cases such as coding should strengthen the case for economic ROI. The institution nevertheless expects debate to intensify around financing availability, datacenter land, power and shell capacity, social and political constraints on new capacity, and the durability of AI capital expenditure amid macroeconomic and interest-rate sensitivity. It expects merchant solutions to remain dominant near term, but ASICs and other specialized accelerators to gain share over time as increasingly diverse workloads require more cost- and power-efficient computing. Within this theme, Nvidia is expected to retain a bullish demand outlook after guiding to 70% CY27 revenue growth, discuss its Rubin ramp and physical AI opportunity, and defend its ecosystem against custom-silicon competition. Goldman Sachs expects AMD to emphasize MI4XX rack-level solutions, ROCm and server-CPU demand as CPUs become more important for agentic AI. Broadcom is expected to stress AI networking and custom XPUs, Qualcomm its datacenter and edge-AI diversification, and Credo ongoing AEC demand plus a progression from Zero-Flap optical transceivers toward ALCs and OmniConnect/Weaver gearboxes. For wafer-fab equipment, Goldman Sachs expects an optimistic 2027 outlook as CapEx estimates rise, driven by DRAM, leading-edge foundry and advanced packaging. It expects about 35% WFE growth in 2026 and sees momentum continuing through at least 2028, with growth accelerating in 2027. The expected spending ramp is capacity-led in DRAM, leading-edge foundry and advanced packaging, with improving investment in NAND and trailing-edge logic. Terafab initiatives and a potential Intel re-acceleration are expected to be debated. The institution considers companies with greater deposition and etch exposure best positioned and highlights Buy-rated Applied Materials and Lam Research; it also sees continued advanced-packaging momentum benefiting Buy-rated Onto Innovation and Qnity. In memory and storage, Goldman Sachs expects management teams to reinforce tight supply-demand conditions through 2027. Its forecasts imply DRAM undersupply of 5.0%, 5.9% and 3.9% in 2026E, 2027E and 2028E, respectively, and NAND undersupply of 4.4%, 4.6% and 3.0%. China-based capacity additions by CXMT in DRAM and YMTC in NAND are a key supply-side uncertainty. The report also flags possible content-per-box rationalization in AI servers, which could reduce memory content in some configurations, and sees capital-return execution—generally announced at 50% to 100% of excess free cash flow—as relevant to valuation. Goldman Sachs expects SanDisk and Seagate to underline limited bit-supply additions over the next 18 months; it remains Buy-rated on SanDisk because of constrained near-term NAND supply, strong demand and improving portfolio mix, and on Seagate because of HDD tightness, pricing and HAMR leadership. For analog, the institution expects positive commentary on demand, revenue visibility, longer lead times, shipments closer to end demand and pricing actions that could more than offset higher input costs. Analog units were roughly 5% above trend as of June, yet Goldman Sachs views the sector as still at the beginning of a recovery because shipments were below trend for a prolonged period over the prior four years and demand improvement is broadening. AI datacenter demand could make the recovery more structural, while automotive demand, inventory restocking and pricing effects remain important questions. It is constructive on NXPI for operational execution and on SITM for AI-datacenter growth plus volume trends in other end markets. For EDA software, Goldman Sachs expects custom AI silicon to increase demand for chip design tools and IP. Cadence is expected to describe long-term EDA and IP growth, driven by more-customized designs, broader customer demand for IP and autonomous-agent offerings that are accelerating revenue growth versus the pre-AI period. Goldman Sachs argues that custom AI silicon worsens a structural shortage of chip-design engineers, creating an Agentic AI monetization opportunity for EDA vendors estimated at about $3.7bn per year by 2030. It says this opportunity is not reflected in Street estimates and may begin to show up as early as 2H26; AI-driven content and pricing tailwinds, including demand in China, are expected to support the outlook.

Analysis framework

Goldman Sachs organizes the preview around investor debates for five semiconductor subsectors. It combines recent company results and management commentary with supply-demand expectations, capital-spending trends, competitive architecture shifts, product roadmaps and company-specific conference expectations to identify likely catalysts and constraints.

Methodology notes

• Industry AnalysisSupply-demand framework

  Memory and storage supply-demand balance

  The report estimates industry undersupply for DRAM and NAND and assesses how capacity additions, demand growth and server-content changes could affect market tightness and company outcomes.

• Industry AnalysisUpstream-Midstream-Downstream Transmission

  AI infrastructure investment transmission through the semiconductor supply chain

  The analysis links hyperscaler AI spending and datacenter capacity needs to demand for compute, networking, equipment, memory, storage, analog and EDA products.

• Industry AnalysisVolume-price decomposition

  Analog volume, pricing and margin dynamics

  The report distinguishes improving unit shipments and end-market demand from pricing actions and input costs to explain potential revenue and margin effects.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Nvidia

  Expected beneficiary of robust AI infrastructure demand and both open and proprietary model ecosystems.

  Strengths

  Recent guidance for 70% CY27 revenue growth; Rubin ramp, physical AI opportunity and ecosystem strategy.

  Comparison

  Expected to address competition from custom-silicon vendors.

  Risks

  AI CapEx durability, datacenter capacity constraints and competitive architecture shifts.

• AMD

  Expected to benefit from expanding datacenter AI and growing CPU relevance in agentic AI workloads.

  Strengths

  MI4XX rack-level solutions and ROCm software stack.

  Comparison

  Seeking to improve its competitive position in datacenter AI.

  Risks

  Competitive dynamics in AI accelerators and execution on rack-level solutions.

• Applied Materials

  Highlighted as a WFE-upturn beneficiary through deposition and etch exposure.

  Strengths

  Expected relative strength across foundry, DRAM and advanced packaging.

  Comparison

  Goldman Sachs expects potential market outgrowth.

  Risks

  Duration and composition of the WFE upturn.

• Lam Research

  Highlighted as a WFE-upturn beneficiary through deposition and etch exposure.

  Strengths

  Potential share expansion in DRAM, foundry and NAND.

  Comparison

  Management may discuss outgrowing the market and reaching a 55% gross-margin target.

  Risks

  WFE-cycle duration and target-achievement timing.

• SanDisk

  Buy-rated beneficiary of NAND tightness and AI-inference storage demand.

  Strengths

  Limited near-term NAND supply additions, improving product portfolio and mix, LTAs and HBF roadmap.

  Comparison

  Expected to discuss demand and constrained supply alongside Seagate.

  Risks

  NAND supply additions and potential AI-server memory-content rationalization.

• Seagate

  Buy-rated beneficiary of HDD tightness and pricing.

  Strengths

  Persistent HDD tightness, strong pricing trends and HAMR technology leadership.

  Comparison

  Expected to discuss demand and constrained supply alongside SanDisk.

  Risks

  Demand conditions, technology-ramp execution and pricing expectations.

• Cadence

  Goldman Sachs remains bullish on Cadence as custom AI silicon and autonomous-agent adoption expand EDA demand.

  Strengths

  EDA and IP exposure, autonomous agents and potential Agentic AI monetization.

  Comparison

  Expected to benefit from a structural chip-design-engineer shortage.

  Risks

  Realization timing of AI-driven revenue and pricing tailwinds.

• NXP

  Goldman Sachs is constructive on the analog recovery through operational execution.

  Strengths

  Secular growth drivers and automotive demand trends.

  Weaknesses

  No expected evidence of customer inventory restocking; price action expected to be neutral.

  Risks

  Automotive recovery timing and inventory dynamics.

• SiTime

  Goldman Sachs is constructive on growth momentum in AI datacenters and other end markets.

  Strengths

  Datacenter demand, smartphone drivers and potential synergies with the acquired Renesas timing portfolio.

  Risks

  Execution on portfolio synergies and end-market demand.

Key data

• Nvidia CY27 revenue growth guidance70%Recent guidance referenced as support for an expected bullish demand outlook.
• Expected WFE growth in 2026~35%Goldman Sachs expects 2027 growth to potentially accelerate beyond this level.
• DRAM industry undersupply5.0% / 5.9% / 3.9%Goldman Sachs forecasts for 2026E / 2027E / 2028E.
• NAND industry undersupply4.4% / 4.6% / 3.0%Goldman Sachs forecasts for 2026E / 2027E / 2028E.
• Capital-return plans50% to 100% of excess free cash flowGeneral range cited as relevant to memory and storage valuations.
• Analog units versus trend~5% above trendAs of June, based on the report's exhibit.
• EDA Agentic AI opportunity~$3.7bn/year by 2030Goldman Sachs estimate; may start to become evident as early as 2H26.
• Lam Research gross-margin target55%Management may indicate the timeframe for reaching the new target.

Impact & implications

The report argues that AI infrastructure demand is broadening the semiconductor opportunity beyond GPUs into CPUs, custom silicon, networking, optical interconnects, memory, equipment and EDA. At the same time, the durability of investment depends on funding, physical datacenter capacity, supply additions, architecture shifts and macro-sensitive risk appetite.

Risks

• AI infrastructure spending could be constrained by financing availability, datacenter land, power and shell capacity, and political or social restrictions.
• Investor risk appetite for AI capital expenditure remains sensitive to the macroeconomic and interest-rate backdrop.
• Custom silicon, ASICs and specialized accelerators could alter competitive positioning from the current merchant-solution-dominated architecture.
• China-based DRAM and NAND capacity expansion could affect memory supply-demand balance.
• AI-server content rationalization could reduce memory content in certain configurations.
• The duration and mix of the WFE upturn, including customer initiatives and Intel’s trajectory, remain debated.

What to watch

• Management commentary on hyperscaler spending, agentic-AI adoption, AI infrastructure ROI and datacenter capacity heading into 2027.
• WFE growth expectations for 2027 and the relative strength of DRAM, foundry, NAND, advanced packaging and trailing-edge logic.
• Memory supply additions from CXMT and YMTC, capital-return execution and the adoption of HAMR and HBF technologies.
• Analog lead times, shipments relative to end demand, automotive recovery, inventory dynamics and pricing effects.
• Evidence that EDA autonomous-agent products are monetizing and that AI-driven content and pricing tailwinds are accelerating revenue.