← signals
2026-07-24·MSFT·partnership
highup

Microsoft announced a major enterprise AI deployment with AT&T using Microsoft Foundry and AMD GPUs.

Microsoft announced a major enterprise AI deployment with AT&T using Microsoft Foundry and AMD GPUs.

window 60devidence 15confidence score 100price MSFT $401.10

confidence score

Strong evidence: 10 independent source classes support this read.

100
high confidence10 independent source classesofficialothernewsmarketcommunitypasses publish gate
priced-in check

MSFT has not made a large direction-matching 30-90 day move yet.

not priced in
as of 2026-07-167d n/a45d n/a90d -5%yahoo

signal brief

Microsoft announced a major enterprise AI deployment with AT&T using Microsoft Foundry and AMD GPUs. AT&T built its OTel2.0 telecom AI models on Foundry Managed Compute, processing trillions of tokens across multiple open models (Phi-4, OSS-120B, Gemma-4). Notably, AT&T used approximately 530 GPUs, including 430 AMD Instinct MI300X GPUs, demonstrating heterogeneous compute for cost and performance optimization. This partnership signals growing enterprise adoption of Azure AI infrastructure and validates AMD's GPU competitiveness in large-scale AI workloads. The announcement underscores Microsoft's strategy to offer flexible GPU choices (AMD and NVIDIA) in its cloud platform, potentially strengthening its position against competitors like AWS and GCP.

What the sources said

"AT&T used approximately 530 GPUs through Microsoft Foundry Managed Compute spanning multiple GPU architectures including 430 AMD Instinct™ MI300X GPUs." — Microsoft Azure Blog

"This flexibility illustrates a broader trend across AI development. Organizations increasingly need platforms that allow them to choose the right model for the job, optimize for cost and performance, and scale workloads without rebuilding operational environments." — Microsoft Azure Blog

"Every company in the world needs to build its own AI, and that is only possible with open models and open source." — Jeff Boudier, VP Product, Hugging Face (quoted in Microsoft blog)

The deployment highlights Microsoft's ability to attract large enterprise workloads, potentially impacting competitors like Amazon Web Services and Google Cloud. The use of AMD GPoys also benefits AMD by showcasing competitive performance in real-world applications.

source data used

spillover entities

AMDNVDA

Decision support, not stock advice. This signal is research with cited evidence — not a recommendation to buy, sell, or hold any security.