New Delhi: Anthropic’s Claude artificial intelligence assistant is reportedly seeing a decline in usage at some major technology companies as businesses increasingly develop and rely on their own in-house AI tools. The shift highlights a growing challenge for AI companies that initially benefited from strong enterprise demand but now face customers building customised systems internally.

According to a report by The Information cited by India Today, Claude’s internal usage at Meta and Microsoft has fallen, while both companies are increasing their reliance on internally developed artificial intelligence tools. The development comes as technology giants seek greater control over their AI systems, data and costs. (indiatoday.in)

Meta reduces reliance on Claude

Meta has reportedly been reducing its internal use of Claude while encouraging employees to use its own AI tools.

The company has invested heavily in developing its artificial intelligence ecosystem, including large language models and AI-powered products across its social media platforms. Building more AI capabilities internally gives Meta greater control over how models are integrated into its existing systems.

The shift also comes as Meta continues to compete aggressively in the AI race, investing in computing infrastructure, research and talent.

According to the report, Claude’s usage among Meta employees has declined as the company has promoted its own AI tools for internal tasks. (indiatoday.in)

Meta’s move illustrates a broader trend among large technology companies: rather than depending entirely on third-party AI assistants, companies are increasingly trying to develop systems tailored to their own employees and workflows.

Microsoft also shifts towards internal AI tools

Microsoft is another major company reportedly reducing its reliance on Claude for certain internal use cases.

The software giant has invested billions of dollars in AI infrastructure and has integrated its own Copilot products across Windows, Microsoft 365, GitHub and other services.

As Microsoft’s internally developed AI capabilities become more advanced, the company has more opportunities to use its own technology for coding, productivity and enterprise workflows.

The reported decline in Claude’s internal usage does not necessarily mean Microsoft has completely abandoned Anthropic’s technology. Large companies often use several AI models simultaneously, selecting different systems depending on performance, cost, security and the specific task involved.

However, the reported shift indicates that in-house AI is becoming increasingly important even among companies that have access to leading external models. (indiatoday.in)

Why companies are building their own AI

The growing preference for internally developed AI systems is driven by several factors.

One major consideration is data control. Companies handling sensitive information may prefer systems that can be deployed within their own infrastructure or under tighter controls.

Another factor is cost. Running AI services at large scale can become expensive, particularly when employees use AI assistants for coding, research and other computationally intensive tasks. Developing or deploying internal models can potentially reduce long-term dependence on external providers.

There is also the issue of customisation. An internal AI tool can be designed around a company’s specific software, data, processes and security requirements.

For large technology companies with substantial AI engineering teams, building such systems may increasingly be a practical alternative to relying exclusively on third-party assistants.

Claude remains important in the AI market

Despite the reported decline in internal usage at Meta and Microsoft, Anthropic remains one of the leading companies in generative AI.

Claude has gained popularity particularly among developers and enterprise users, with its models being used for coding, analysis, writing and other professional tasks.

Anthropic has also developed partnerships and integrations with major technology platforms. Its models are available through cloud and enterprise channels, allowing businesses to incorporate Claude into their own applications.

The company’s challenge is therefore not simply about maintaining consumer usage. It must also convince enterprise customers that its models offer enough performance, reliability and value to justify continued use alongside increasingly capable internal systems.

AI competition moves beyond model performance

The reported changes at Meta and Microsoft underline how competition in artificial intelligence is evolving.

Earlier in the generative AI boom, the primary focus was on which company had the most capable large language model. Today, businesses are increasingly evaluating the entire AI stack, including infrastructure, specialised models, agents, data access, security and integration with existing software.

For companies such as Meta and Microsoft, developing their own AI tools can also provide strategic advantages. They can integrate models directly into products, customise them for specific workloads and decide how data is processed.

At the same time, external AI companies such as Anthropic can continue to play an important role by providing specialised models that may outperform internal systems for particular tasks.

This could result in a more fragmented enterprise AI landscape, where companies use a combination of proprietary models and third-party systems rather than relying on a single provider.

What the shift means for Anthropic

For Anthropic, the reported decline in Claude’s internal usage at two major technology companies is a reminder of the increasingly competitive nature of the AI industry.

The company must continue improving model capabilities while demonstrating clear benefits to enterprise customers. Its ability to maintain strong developer adoption and secure long-term business partnerships will be particularly important as customers gain more options.

The development also shows that even major AI companies cannot assume that enterprise users will remain dependent on external models indefinitely.

As AI tools become easier for businesses to develop, customise and deploy, the next stage of the industry could see a greater balance between third-party AI providers and proprietary systems.

For Anthropic, Claude’s continued growth will depend not only on the quality of its models but also on how effectively it can remain valuable in a market where some of its biggest customers are increasingly capable of building AI tools of their own.