Trump's Stargate for dummies

I work for Eviden, a multi-billion-dollar consultancy and Global Systems Integrator (GSI) specializing in helping clients utilize AI on AWS, GCP, Azure, Oracle Cloud Infrastructure (OCI), and eventually Stargate. Our expertise spans across industries, helping organizations adopt AI solutions responsibly and effectively.

Disclaimer: These are my personal viewpoints and perspectives, which may differ from those of Eviden or other experts in the field. Nonetheless, I feel compelled to summarize what the Stargate announcement means for AI and the broader tech ecosystem.

What is Stargate? Private Cloud? Public Cloud? Something in Between?

The Stargate AI infrastructure project, announced by President Trump, is likely somewhere between a private and public cloud. Unlike general-purpose cloud providers like AWS or Azure, Stargate is specialized in AI. It won’t be a full-fledged cloud service competing across the board, but rather a focused infrastructure provider for AI workloads.

Delivering AI solutions requires a full stack—compute power, data pipelines, security, and governance. While Stargate will provide high-performance computing (HPC) and AI model hosting, businesses still need complementary cloud, networking, and software platforms.

Credible references, such as the Gartner Hype Cycle for AI and IDC reports on AI infrastructure trends, suggest that hybrid and specialized AI clouds are an emerging category. AI-focused data centers are increasingly being deployed for dedicated workloads rather than as part of traditional cloud environments.

SoftBank’s Investment – $100B in Year One

One of the most notable aspects of Stargate is that SoftBank, a Japanese investment powerhouse, is contributing $100 billion in the first year. SoftBank has a history of making massive tech bets, and this investment signals their intent to own a share of the U.S. AI infrastructure market.

Their return on investment (ROI) will come from offering AI computing services to U.S.-based companies, similar to how cloud providers monetize infrastructure.

This is a real estate and talent acquisition play, two areas that Trump claims to be an expert in. Owning AI real estate means controlling land, power, and physical infrastructure, which is a different strategy than traditional software-based AI investments.

The Reality of AI Data Centers – 10-20 in the Works?

The announcement suggests that between 10-20 data centers are already in development.

How Data Center Real Estate Works

• Land Selection – Data centers require cheap land, proximity to major network backbones, and low disaster risk (e.g., avoiding flood zones).
• Cooling Challenges – AI workloads generate massive heat, requiring advanced liquid cooling systems. Many hyperscale AI data centers are turning to immersion cooling to handle the environmental constraints.
• Water Constraints in Texas – Many of these data centers are planned for Texas, which is already struggling with water supply issues. AI data centers consume large amounts of water for cooling, which could become an issue for scalability.

Energy & Power Contracts

• AI data centers need long-term energy contracts to secure high power loads.
• Trump's opposition to solar and wind suggests alternative energy sources will be prioritized.
• Conflicts with Green Initiatives? Traditional data center operators like AWS, GCP, and Azure have committed to renewable energy goals, while Stargate’s strategy remains unclear.

Stargate vs. Public Cloud AI Services

Stargate introduces a new layer of competition for AI cloud services, particularly against AWS, GCP, and Azure.

• Oracle’s AI Capabilities in OCI are Limited – Oracle, a key Stargate backer, offers AI services, but they lag behind AWS, Azure, and Google in AI model hosting.
• OpenAI’s Relationship with Microsoft – OpenAI is heavily invested in Azure’s cloud infrastructure. If Stargate is hosting OpenAI models, does this mean Microsoft will power these data centers?
• AWS & Google’s AI Investments – AWS and Google are both aggressively expanding AI-focused data centers. It’s unclear how Stargate will compete or differentiate itself.

Nvidia’s Role in Stargate

Nvidia was explicitly mentioned, implying GPUs will play a significant role. Given Nvidia dominates AI chip production, it’s not surprising that Stargate will rely on Nvidia hardware. However, it’s still unclear what mix of chips, compute, and cloud services will be offered.

The Impact on ISVs and Open Source AI Contributors

Independent software vendors (ISVs) and open-source AI contributors may be affected by this shift. Some key questions:

• Will open-source AI models be hosted on Stargate?
• How will ISVs integrate with this infrastructure?
• Will consultancies and AI service firms be involved?

Eviden is already working with AI in Texas, such as our collaboration with the University of Texas on opioid crisis research (case study here). Whether consulting firms will play a role in Stargate’s AI ecosystem is still an open question.

How Does This Compete with U.S. Tech Giants?

While Trump and Stargate mention medical applications, AI has far broader use cases, including finance, logistics, cybersecurity, and defense. This raises concerns about competition with U.S. tech giants.

It’s also not uncommon for governments to invest in AI. Examples include:

• DARPA’s AI funding for defense applications.
• National AI Research Institutes, backed by NSF, focusing on ethical and scalable AI.

Final Thoughts – Uncertainty & AI’s Future

While Stargate raises many questions, as an AI expert, I remain focused on solving real-world problems. AI infrastructure is an enabler, but it’s not the only factor in AI success.

The overlap between AI, space exploration, and quantum computing suggests that infrastructure alone is not enough—we need holistic AI strategies.

Additionally, Deloitte’s recent AI report (read here) highlights that AI challenges extend beyond infrastructure to include responsible AI, ethics, and governance.

As Stargate develops, the AI community will need to adapt, evaluate, and integrate responsibly. The road ahead is uncertain, but one thing is clear: AI is here to stay.