2 July 2026

 

Globant announced on 30 June that it has entered into a multi-year alliance with Anthropic, becoming a Preferred Services Partner within the Claude Partner Network and integrating Claude-powered AI Pods into its enterprise offerings. The agreement gives 28,500 Globant employees access to Claude certifications and expands the company's capabilities around enterprise artificial intelligence deployments.

At first glance, the announcement resembles another partnership between a foundation model developer and a global consultancy. Yet it also raises a larger question about the direction of enterprise AI.

Are we witnessing the emergence of the AI equivalent of Windows, SAP or AWS?

For much of the past three years, competition in artificial intelligence has been framed around models. OpenAI's GPT family, Anthropic's Claude, Google's Gemini and a growing number of specialised systems have competed on benchmarks, coding performance, reasoning capabilities and multimodal functions.

Enterprise technology history, however, suggests markets do not always reward the strongest individual product. Companies that built extensive ecosystems of developers, consultants, resellers and implementation partners often created advantages that persisted long after products evolved or competitors caught up.

Anthropic's recent announcements suggest the company increasingly sees enterprise adoption through that lens.

 

From model provider to platform builder

 

Anthropic launched the Claude Partner Network in March 2026, committing $100 million to certifications, technical enablement, partner training and go-to-market support for organisations deploying Claude within enterprises.

In June, the company expanded the initiative with a Services Track and a Partner Hub designed to connect customers with certified implementation specialists. According to Anthropic, the programme has already attracted more than 40,000 applications and produced over 10,000 consultant certifications within months of launch.

The structure of the programme also appears designed for scale. Anthropic introduced tiered designations, including Select, Preferred and Premier partner levels. Preferred Services Partners are expected to demonstrate significant implementation experience, including customer deployments and a sizeable base of certified practitioners.

Globant's entry into the programme adds another large services company to the network.

The Luxembourg-headquartered technology firm, founded in Argentina and employing more than 30,000 people worldwide, plans to incorporate Claude into its AI Pods offering, specialised teams designed to combine sector expertise with agentic AI capabilities.

Globant says AI Pods are already used by 40 per cent of its twenty largest customers, indicating that enterprises are beginning to move beyond experimentation and into operational deployments.

That reflects a challenge facing many organisations.

Few enterprises purchase language models as standalone products. Most require implementation expertise, governance frameworks, workflow integration, security controls and ongoing operational support. Deploying AI at scale often means connecting models with existing applications, internal data systems and business processes.

Those activities typically fall to systems integrators, consultants and specialised technology partners.

Anthropic appears to be investing heavily in that layer.

 

Enterprise technology has rewarded ecosystems before

 

The idea of competing through ecosystems rather than products is not new.

Microsoft transformed Windows from an operating system into a platform supported by resellers, software developers, independent vendors and systems integrators. SAP expanded globally through consulting firms capable of embedding enterprise software into large organisations. AWS built an extensive network of cloud specialists, managed service providers and software partners.

Salesforce offers another example.

The company has issued more than one million certifications across its ecosystem and supports thousands of consulting partners delivering projects worldwide. According to research commissioned by Salesforce and conducted by IDC, the Salesforce ecosystem is expected to generate $1.6 trillion in economic activity by 2026.

These networks were not created overnight.

They emerged over years as developers, consultants and businesses identified commercial opportunities in helping customers adopt new technologies.

Anthropic remains considerably smaller in scale.

Its approximately 100 launch partners and more than 10,000 certified consultants represent only a fraction of Microsoft's decades-old partner organisation or Salesforce's mature consulting ecosystem. Yet the pace of growth suggests Anthropic sees enterprise adoption as more than a product challenge.

Since March, the company has allocated substantial resources towards certifications, technical support, implementation expertise and partner enablement. The objective appears to be creating conditions in which third parties can build businesses around Claude.

Enterprise technology history contains numerous examples of companies whose ecosystems eventually became as influential as the products themselves.

 

Beyond the model wars

 

The focus on benchmarks and leaderboards has been understandable.

Generative AI has advanced at remarkable speed, with model developers regularly releasing new versions promising improved reasoning, coding and multimodal capabilities.

At the same time, enterprises are increasingly asking different questions.

How quickly can AI systems be integrated into existing workflows?

Who is responsible for implementation?

How are security and governance managed?

What happens when systems require updating, monitoring or adapting over time?

The answers increasingly involve organisations beyond the model developers themselves.

Microsoft enters this phase with substantial advantages through Azure, Microsoft 365, GitHub and long-established enterprise distribution channels.

Google combines Gemini with cloud infrastructure, productivity software and hardware platforms.

Salesforce is embedding autonomous agents into customer relationship management systems through Agentforce.

OpenAI continues to expand its enterprise footprint through APIs, software integrations and partnerships.

Anthropic, meanwhile, is investing in a services ecosystem designed to encourage consultants, developers and implementation specialists to participate in Claude's growth.

Members of the Claude Partner Network receive access to engineering support, certifications, training resources and co-selling opportunities.

Rather than positioning Claude solely as a standalone service, Anthropic is building many of the structures traditionally associated with enterprise platforms.

 

Europe has its own contender

 

For Europe, the emergence of ecosystem competition presents a different challenge.

The continent has produced globally relevant AI companies before, but it has often struggled to translate technical expertise into dominant software platforms.

Mistral AI offers perhaps the clearest contemporary example.

Founded in Paris in 2023, the company rapidly established itself as Europe's most prominent foundation model developer, attracting partnerships with cloud providers, enterprises and public-sector organisations while positioning itself as an advocate for open and sovereign AI deployment.

The comparison with Anthropic is revealing.

Anthropic is investing heavily in implementation capacity, certifications and partner enablement. Mistral has concentrated on openness, accessibility and giving organisations greater control over where models are deployed and how data is managed.

Both approaches address enterprise adoption, albeit from different directions.

One seeks to cultivate a network of consultants, integrators and developers around a model. The other emphasises openness, interoperability and technological sovereignty.

History suggests ecosystems can become durable competitive advantages.

Microsoft benefited from decades of relationships with resellers, developers and implementation partners. SAP expanded internationally through consulting firms capable of embedding its software inside large organisations. AWS built a vast community of cloud specialists and software providers that accelerated adoption.

The question for Europe is whether companies such as Mistral can foster comparable ecosystems around their technologies, or whether the next generation of enterprise AI platforms will once again be centred primarily around American firms.

 

Claude as an operating layer

 

Globant's announcement offers a glimpse of what that future may look like.

Its AI Pods initiative is intended to embed Claude inside operational environments rather than use it as an isolated assistant.

That reflects a wider development taking place across enterprise AI.

Generative models are gradually becoming components within larger systems, orchestrating workflows, interacting with applications, retrieving information and supporting decision-making processes.

The transition resembles earlier moments in enterprise technology.

Windows became a platform through which businesses accessed software.

SAP evolved into a foundation layer for corporate processes.

AWS transformed infrastructure into a service consumed through an extensive partner ecosystem.

Anthropic's recent investments suggest the company increasingly views Claude in similar terms.

The question is no longer simply whether Claude can outperform competing models in a benchmark test.

It is whether Anthropic can create the conditions in which consultants, software vendors, developers and systems integrators see Claude as a platform worth building around.

Enterprise technology history offers numerous examples of companies whose ecosystems eventually became more influential than their products alone.

Anthropic is attempting to build one of those ecosystems around Claude. Whether Europe succeeds in creating an equivalent around companies such as Mistral may prove to be one of the defining questions of the next phase of enterprise AI.