
Lisa Su
Chair and CEO · AMD
Joint fourteenth: concrete humanoid computing contributions, with material but not pervasive humanoid developer dependence.
Analysis
Established position
AMD’s January CES announcement identifies Su as chair/CEO, names Generative Bionics among technology partners and explicitly includes humanoids in its embedded-processor applications. This is a humanoid-computing route; general data-centre scale is only an enabling resource, not humanoid market control. AMD — 2026-01-05
Analysis
Strategic thesis
Su can supply alternative training and embedded computing for humanoid developers, creating strategic influence over cost and architectural choice if the named humanoid relationships grow into repeat design adoption with practical software support and acceptable integration effort. Competitive compute and embedded platforms can reduce dependence on one supplier. Software compatibility, developer support and whole-system efficiency determine whether silicon advantages become usable robotics advantages.
Analysis
Explicit assumption
AMD sustains software and embedded investment and robotics teams can adopt its platforms without disproportionate porting or integration costs.
Analysis
Strongest counter-case
Chip competition does not automatically create robot ecosystem control. A hardware roadmap may outpace software readiness, and customer-specific optimisation can favour incumbent tools or custom architectures. Confidence: Humanoid-specific computing contributions are identifiable; adoption and software influence are materially less certain than AMD’s broad computing relevance.
Analysis
Forward scenarios through 2030
Base: AMD supports selected humanoid training and embedded-compute designs. Upside: Improved software and economics make it a widely adopted alternative platform. Downside: Porting friction leaves robot builders concentrated on competing stacks.
Analysis
What to monitor
- 01Robot-related production platform wins.
- 02Software portability and developer maintenance burden.
- 03Delivered embedded/training products versus roadmap promises.
Analysis
Research boundaries
Corporate R&D allocation; no dedicated robotics budget verified. Chair/CEO within public-company governance. The full capital, talent, compute, data, manufacturing, distribution, architecture, integration, dependence and governance record is structured in the snapshot.
AMD embedded AI Ryzen AI P100/X100 programme: announced products; AMD Instinct MI500: 2027 roadmap.
Sources were accessed on September 13, 2026. Undated profiles establish only the current reviewed record; future product, capacity or delivery statements remain plans. Exact budgets, data rights and private delegation are not assumed from the title.
Analysis