🔍 Read the full analysis: How Companies Can Estimate The Cost Of Switching From Claude on ThorstenMeyerAI.com
Get the latest gadgets delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
TL;DR
A report by The Information says Meta and Microsoft have reduced some employees’ use of Claude while directing them to tools they own or back. The figures do not establish that Claude performed worse or that either company has ended access. For other businesses, estimating a switch means counting evaluation, engineering, integration and productivity costs—not just comparing model prices.
A report by The Information on October 5 says Meta and Microsoft have reduced some employees’ use of Anthropic’s Claude tools and directed them toward alternatives they own or back. The reported moves put a practical question before other companies using AI: how to estimate the full cost of switching when the price of a model is only one part of the expense.
The Information reported that Meta cut the number of employees using Claude Code from about 60,000 to about 30,000. The source material says Meta has been steering staff toward its internal tools, MetaCode, which has more than 30,000 users, and Muse Code, with more than 6,000. These are reported figures; the available material does not provide a company statement confirming the numbers or the timing of each change.
Microsoft reportedly had projected more than $1 billion a year in internal spending on Anthropic technology, including Claude Code, Claude models in Copilot and Claude Mythos. It has since cut that projection by more than a third, according to the report, and is steering employees toward GitHub Copilot and OpenAI models. The source also says Microsoft continues to spend on Anthropic models for customer-facing Copilot features, and that customer spending on Claude through Microsoft platforms is growing.
The reported explanations are cost controls and available alternatives, not a stated finding that Claude is lower quality. Microsoft is also said to have imposed tighter token budgets; one account cited in the source says some monthly team budgets fell from around $100,000 to around $10,000. That detail comes from a single report and should not be treated as a company-wide policy. Neither account, as presented, says Claude access has ended.
Meta and Microsoft pulled back from Claude. Here’s what switching actually costs.
The Information reports both companies steering their own employees away from Claude. Read as a verdict on Claude, it misleads. Read as a demonstration of switching — and who can afford it — it’s the most useful enterprise-AI signal this month.
Staff steered to GitHub Copilot and OpenAI models; stricter token budgets. One unconfirmed report: some team budgets ~$100k → ~$10k/month.
Microsoft reportedly still spends heavily on Claude for customer-facing Copilot — and that spending is reported to be growing.
Reported drivers: rising token costs and owned alternatives. Neither company is reported to have called Claude worse.
Meta builds coding tools; Microsoft owns Copilot and backs OpenAI. This is ordinary vertical integration.
Keep a second vendor live on real work.
A few hundred tasks with pass criteria.
Logic, prompts, tools in your layer.
Tokens are the cheap half.
Know what you’d rebuild.
On the evidence reported, Meta and Microsoft didn’t reject Claude. They brought spending in-house where they could and kept buying where they couldn’t — Microsoft remains a large Anthropic customer for the products it sells. The signal is the mechanism: the most sophisticated buyers treat models as interchangeable suppliers behind a layer they control.Meta could halve its Claude usage because it had built somewhere else to go. Build somewhere else to go.
The Costs Behind a Model Switch
The reported shifts show why a company’s AI bill cannot be assessed by comparing token prices or subscription fees alone. A switch can require new evaluations, engineering changes and employee training. It can also affect how much review and rework people need before they accept a model’s output.
Companies with their own alternatives may be able to justify those expenses through lower ongoing costs or tighter control over tools. The source estimates that a reduction of more than a third from Microsoft’s reported $1 billion-plus projection could amount to more than $300 million annually. That is an illustrative calculation based on reported figures, not a confirmed saving or a public accounting of realized costs.
Smaller buyers may face a different calculation. A business spending $20,000 a month could find that engineering and transition expenses outweigh a year of savings, but that outcome depends on its workload, internal capabilities and the quality of alternatives. The point is to compare cost per accepted result, including human review, rather than treating a lower model price as proof of lower total cost.
As an affiliate, we earn on qualifying purchases.
Why Meta and Microsoft Can Move
The companies described in the report are not typical AI buyers: each has access to credible alternatives. Meta develops internal models and coding tools; Microsoft owns GitHub Copilot and backs OpenAI. Directing employees toward products a company owns or supports can reflect competitive strategy and internal cost management, as well as a judgment about usefulness. It does not, by itself, show that an outside product failed.
That distinction matters when applying the report to other organizations. Meta and Microsoft can draw on engineering teams and tools already deployed across their businesses. A company without a second production-ready system may need to build integrations and workflows before it can make a comparable move. The source’s broader recommendation is to prepare for choice in advance: keep more than one model family in use where practical, maintain representative evaluation tasks, and keep prompts and business logic in a layer the company controls.
enterprise AI cost estimation software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What the Report Does Not Establish
The figures and account of internal policy changes are reported claims in the supplied source material; no direct statements from Meta, Microsoft or Anthropic are included. The precise dates, scope and methods behind the employee and spending figures are not specified here. It is also unclear how much of the projected Microsoft spending was actually incurred before the projection changed.
The report, as summarized, does not establish that either company has ended its relationship with Anthropic, that customer access to Claude has changed, or that the tools chosen as alternatives produce better results on the companies’ tasks. It also does not quantify the engineering, review or productivity costs of the internal shifts. Those unknowns limit what can be concluded about the overall savings.
As an affiliate, we earn on qualifying purchases.
Measure Before Changing Providers
Companies weighing a switch can start by recording current spending alongside time spent on review, rework and maintenance. They can then test alternatives on a representative set of real tasks, with clear criteria for acceptable results, before moving more work. The comparison should include integration effort, employee learning time and any changes to cached context or usage pricing.
Keeping a second provider active on a limited share of real work can reduce the cost of building an alternative from scratch later. No future change by Meta or Microsoft, or further public explanation of their reported decisions, is specified in the source material. For now, the practical next step for other buyers is to build their own evidence before treating a price difference as a business case.
AI productivity management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Have Meta and Microsoft stopped using Claude?
The supplied account does not say that either company has ended Claude access. It reports reduced internal use in some areas and says Microsoft continues to spend on Anthropic models for customer-facing Copilot features.
Why are the companies reportedly shifting employees to other tools?
The reported reasons are rising token costs, tighter spending controls and in-house alternatives. The source does not report either company saying Claude performed worse.
What costs should a company include in a switch estimate?
Include evaluation work, prompt and integration changes, staff training, productivity disruption, cache or usage changes, and possible increases in human review and rework—not only subscription or token prices.
How can a company test whether an alternative is suitable?
Run both systems on representative tasks with clear pass criteria, then compare quality and the time required for review and correction. A consistent evaluation set makes the decision easier to measure.
Source: ThorstenMeyerAI.com
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
