AI chips and networking
The companies designing the processors, accelerators and networking equipment that do the actual computing behind AI, including GPUs like NVIDIA's, and the high-speed connections that link thousands of them together.
Every connection below is tied to a readable source, date, classification and statement of its limits. A company source establishes what that company stated. It does not necessarily provide independent confirmation.
Cisco Systems partnering on rack-scale Secure AI Factory systems with NVIDIA Supermicro
Cisco expanded its NVIDIA-based Secure AI Factory offering to include Supermicro as a rack-scale systems partner, combining Cisco networking with Supermicro's liquid-cooled AI servers.
Source: 2026-08-25 · reviewed 2026-09-20
Evidence: Primary-source statement · Newly announced, current three-way go-to-market partnership
Limits: Confirms a joint go-to-market and technical partnership; does not disclose revenue, deployment volumes, or customer commitments.
Cisco Expands Secure AI Factory with NVIDIA for the Rack-Scale Era ↗
NVIDIA supplies GPUs to Starcloud
Starcloud said its Starcloud-1 satellite, launched in November 2025, carried the first data-center-class NVIDIA H100 GPU to operate in orbit.
Source: 2026-08-21 · reviewed 2026-09-24
Evidence: Primary-source statement · Reported operating
Limits: Establishes that one NVIDIA GPU flew on one satellite. It does not establish the commercial terms, whether the part was sold, donated or supplied through a program, any volume commitment, or that NVIDIA will supply GPUs for a larger constellation.
Starcloud Raises $250 Million at $2.3 Billion Valuation to Scale AI with Orbital Data Centers ↗
Marvell Technology co-develops custom silicon with Google (Alphabet subsidiary)
Marvell's SEC 8-K discloses a commercial agreement, signed July 29, 2026, for custom silicon programs that attach to Google's TPU ecosystem, including AI inference accelerators, storage controllers, network interface controllers, memory interface controllers and near-memory compute products.
Source: 2026-08-19 · reviewed 2026-09-22
Evidence: Primary-source statement · Reported operating
Limits: Describes the scope of custom-silicon products covered by the agreement; does not establish current shipping volumes or which products have already been delivered.
VIAVI supplies test and measurement products to Nokia
VIAVI's fiscal 2026 annual report lists Nokia among its customers.
Source: 2026-08-13 · reviewed 2026-09-26
Evidence: Primary-source statement · Reported operating (fiscal 2026)
Limits: A customer-list mention; does not establish which products Nokia buys, revenue from Nokia, or whether the purchases relate to AI hardware.
Viavi Solutions Inc. Form 10-K for fiscal year ended June 27, 2026 ↗
Tenstorrent joined the High Bandwidth Flash consortium with Sandisk
Sandisk's release on the High Bandwidth Flash specification said Tenstorrent joined as a consortium member during the HBF standardization work that began in February 2026.
Source: 2026-08-03 · reviewed 2026-09-26
Evidence: Primary-source statement · Consortium membership reported
Limits: Consortium membership only; does not establish that Tenstorrent buys, has committed to buy, or will deploy HBF products from Sandisk.
AMD tests agentic EDA workflows from Synopsys
AMD is evaluating new Synopsys agentic AI EDA workflows delivered via Microsoft Discovery, including autonomous debug-closure and implementation tools, to accelerate design of next-generation AMD silicon.
Source: 2026-07-27 · reviewed 2026-09-20
Evidence: Primary-source statement · Active collaboration announced; AMD described as testing/evaluating the workflows, not yet full production deployment
Limits: Confirms AMD is testing these specific agentic workflows with an on-record AMD executive quote; does not establish exclusivity, contract value, or that AMD uses Synopsys tools for all of its chip designs.
Synopsys Advances Agentic AI Chip Design with AMD and Microsoft ↗