Detailed Analysis
A verified Reddit user managing a Claude Team Plan account has reported a significant onboarding failure affecting new user provisioning, in which phone number verification requirements are blocking legitimate enterprise team members from completing account creation. The situation involves a sixth user being added to an active, domain-verified team plan — one where the organization has already been billed — yet the new user is caught in a loop where every phone number attempted, including entirely new ones never previously associated with Anthropic, returns the error: "That phone number was used too recently. Try again in 40 hours." Attempts to circumvent the issue through email alias re-invitations and alternative phone numbers produced identical results, suggesting the block is not tied to a specific number or email address but may be linked to the account, device, or network identity of the new user.
Anthropic's official support documentation confirms that phone number verification is a mandatory step for all new Claude account creation, functioning as an identity checkpoint via one-time passcodes sent by SMS. The documentation also acknowledges that the "phone number already used" error typically arises when another Claude account in the system was previously verified with the same number, requiring support intervention to unlink it. However, the scenario described in this report appears distinct: multiple different phone numbers, none previously registered with Anthropic, are all triggering the same 40-hour cooldown error. This points to a possible rate-limiting or fraud-detection mechanism that is flagging the account creation attempts themselves — not the phone numbers — based on behavioral or metadata signals such as IP address, device fingerprint, or the pace of provisioning activity.
The support experience compounds the technical failure significantly. The reporting user describes reaching only an AI support agent that terminates the conversation when the specific error is explained, with no pathway to escalate to a human representative. This represents a structural gap in Anthropic's enterprise support model: a paying Team Plan customer, mid-provisioning on a billed seat, has no reliable means of resolving a system-level block. The inability to bypass or override the verification loop through standard administrative actions — re-invitation, email aliasing, alternate phone numbers — further suggests that Team Plan administrators lack the elevated account management controls that enterprise customers would reasonably expect.
This incident sits within a broader and increasingly visible tension in the AI industry between consumer-grade account security mechanisms and the operational demands of business customers. Phone number verification, a common anti-abuse tool borrowed from consumer platforms, creates friction that is acceptable at the individual sign-up level but becomes a liability when applied rigidly to managed, pre-authorized enterprise provisioning flows. Competitors such as OpenAI and Google have faced similar criticism when consumer-facing identity guardrails have been applied without modification to their enterprise tiers. For Anthropic, which has been actively expanding its Team and Enterprise plan offerings — including the recent addition of Claude Code to those tiers — incidents like this carry reputational weight, as they directly challenge the company's positioning of Claude as an enterprise-ready platform.
The report is unlikely to represent an isolated edge case. The combination of domain verification, active billing, and multiple failed phone numbers across a provisioned account suggests a systemic flaw in how the verification pipeline interacts with team-managed accounts. Until Anthropic either exempts Team Plan provisioning flows from standard consumer phone verification requirements or provides administrators with tools to manually authorize new users, similar blocks will continue to surface as organizations scale their deployments. The absence of a human support escalation path for enterprise customers in this situation is perhaps the most operationally damaging element of the reported experience, and one that Anthropic will need to address as it competes for the enterprise market share that demands both reliability and responsive account management.
Read original article →