Tata Consultancy Services reported on July 9 that its AI business had reached a $2.6 billion annualized revenue run rate in the first quarter of its 2027 fiscal year, up 13.6% from the prior quarter.

The result followed a June partnership with Anthropic under which TCS created a dedicated business unit for Claude and committed to equipping 50,000 associates across engineering, finance, legal, marketing and sales with the models.

The model is only the beginning

Frontier labs can provide capable models and developer platforms. Large enterprises still need to connect those systems to existing software, redesign workflows, manage access, satisfy regulators and train employees. That implementation burden is familiar territory for global consulting and systems-integration firms.

TCS said its Anthropic unit will develop joint industry solutions and take them to market in regulated sectors including financial services, healthcare, aviation and telecommunications. It also plans to contribute reusable Claude Code skills and plugins for domain-specific work such as claims adjudication and lending advice.

Those are company plans, and the $2.6 billion figure covers TCS’s broader AI business rather than revenue attributable only to Anthropic. The two announcements should therefore be read together as strategy and scale—not as proof that the dedicated unit generated that run rate.

Consultancies are becoming distribution

The arrangement illustrates how frontier-model companies can reach large organizations without building every layer of industry expertise themselves. A consulting partner brings executive relationships, existing contracts and staff who understand the legacy systems into which an agent must fit.

That relationship can accelerate adoption, but it can also deepen dependency. A company may become tied to a model provider, a consultancy’s implementation and a set of custom workflows at the same time. Buyers should insist on clear ownership of integrations, evaluation data and model-specific components, along with a plan for switching technology when needed.

The commercial opportunity extends well beyond license resale. The larger revenue pool may include data work, governance, security, process redesign, managed operations and ongoing evaluation. Those services persist as models change.

TCS’s numbers show that the implementation layer is already becoming a substantial business. The frontier labs may define what models can do. Consulting giants are organizing entire units around the harder question: how a large institution actually changes enough to use them.

Sources for editorial review

Drafting note: This draft was prepared with AI assistance from the linked source material and requires author review, independent fact-checking and final editorial approval before publication.