Anthropic and OpenAI Rewrote Their Plans. Steal the Playbook
Anthropic renamed its Claude plans and removed the five-hour limit for verified Ultra accounts, while OpenAI will share one token allowance across ChatGPT and Codex from December. The real lesson is that the limit, not the price tag, is what customers buy.
Key takeaways
- Anthropic renamed Max 5x to Max and Max 20x to Ultra at the same prices, and verified Ultra accounts lost the five-hour reset.
- Claude pools web, desktop, mobile and Claude Code usage, and paid plans add weekly limits on top of the rolling five-hour window.
- OpenAI will share one token allowance across ChatGPT and Codex from December, with credits as the overflow.
- OpenAI also blocks new subscriptions tied to too many cards or devices, and buying credits will not clear that restriction.
- The playbook: name tiers by position, pool usage, sell overflow credits and gate abuse before it costs you money.
Anthropic renamed its Claude plans on 30 October and removed the five-hour limit for verified Ultra subscribers. OpenAI will share one token allowance across ChatGPT and Codex from December. Both look like billing housekeeping. They are really a pricing lesson: the limit, not the sticker price, is what the customer buys.
TL;DR
- Anthropic renamed Max 5x to Max and Max 20x to Ultra at the same prices, and verified Ultra accounts lost the five-hour reset.
- Claude pools web, desktop, mobile and Claude Code usage, and paid plans add weekly limits.
- OpenAI will share one token allowance across ChatGPT and Codex from December, with credits as overflow.
- OpenAI also blocks new subscriptions tied to too many cards or devices, and buying credits will not clear it.
Disclosure: ZeroRaised publishes this analysis and is a launch site for bootstrapped products. Facts come from the linked sources as checked on 2 Nov 2026; ZeroRaised has no commercial relationship with the companies named.
Pooled usage is a pricing design where several surfaces share one metered allowance, so activity in one place reduces what is left everywhere else.
What did Anthropic actually change?#
Anthropic renamed the $100 Max 5x plan to Max and the $200 Max 20x plan to Ultra, kept the prices, and announced it on X on Friday 30 October. In Europe the plans cost €90 and €180 a month (The Next Web).
Verified Ultra accounts no longer have a limit that resets every five hours. Newer accounts keep a temporary limit until Anthropic verifies them, a step it said helps prevent fraud and abuse, and they get an email when it is lifted. Weekly limits stay the same (The Next Web). Claude's official X account told subscribers: "If you're a subscriber, there's nothing you need to do."
The pricing page lists Max at $100 a month for five times Pro usage every five hours and every week, and Ultra at $200 for everything in Max plus no five-hour limit for verified accounts, ten times Pro usage every week and $200 a month in API credits (Claude pricing).
The old Max 20x name pointed to 20 times Pro usage; the page now lists ten times a week, and Anthropic's post did not mention it, according to The Next Web. Separately, a lawsuit filed in June claimed Anthropic oversold usage on both Max plans. That is a claim, not a finding.
Activity across web, desktop, mobile and Claude Code draws from the same pool, and the FAQ says "there's no fixed message count" (Claude pricing).
Why did OpenAI add card and device limits?#
If too many subscriptions sit on the same payment card or device, ChatGPT may block a new subscription purchase, and it limits how many paid accounts one device can use, with the warning "Device limit reached" (OpenAI Help Center).
The rules cover ChatGPT Go, Plus, Pro 100, Pro 200 and Pro 500. Free plans and Business, Enterprise and Edu workspaces do not count. Signing out does not guarantee that another paid account becomes eligible, and buying credits or resetting usage will not clear the restriction (OpenAI Help Center).
OpenAI points users to credits on an existing account, usage resets, cancelling and refunding, or Support for a genuine mistake such as a shared household card. Credits pay for eligible usage after included limits are reached, but they do not combine subscriptions or move usage between accounts. Subscriptions bought before the rules took effect continue under an existing-subscription exception (OpenAI Help Center).
Our read: that exception is the tell. The target is not a normal customer; it is the person running several cheap accounts on one phone.
What happens when ChatGPT and Codex share one allowance?#
From December, the usage allowance on Plus, Pro and Business plans will be shared across ChatGPT and Codex. Usage is measured in tokens, advanced models and complex tasks use more, and "Instant chat will not draw from your usage" (OpenAI Help Center).
OpenAI expects most users will not need to wait for resets more often than now, and it shows combined remaining usage and reset times in Settings > Usage. At the limit, a user can wait, upgrade or buy credits. The Decoder reads this as a hit on people rotating several paid accounts on one device, with OpenAI pointing them to credits or resets, "options that are generally more expensive than an additional subscription" (The Decoder). That is analysis, not a quote from OpenAI.
What is really being sold here?#
Both companies are moving from "how much can I use?" to "what happens when I run out?". A fixed allotment invites a spreadsheet. A pooled allowance with a clear overflow path invites a decision: keep working, and pay for it. Anthropic sells that path as usage credits at standard API rates, and OpenAI sells credits.
The naming carries a second lesson. Max and Ultra describe a position, not a number. The old Max 20x name promised a multiple; the new name promises a tier. If your plans are called Starter 1,000 and Pro 10,000, you have bolted your brand to a number you will want to change.
How should you copy the playbook?#
- Name tiers by position, not quota. A name that survives a usage change is worth more than a precise one.
- Pool usage across surfaces. One allowance is easier to explain than three.
- Sell overflow, do not just block. Credits at your standard rate keep a heavy user in the product.
- Verify new accounts instead of pricing everyone out.
- Say what changed and why. An unexplained gap between a 20x name and a 10x listing is how trust breaks.
What this means for bootstrapped founders#
You do not have Anthropic's fraud problem, but you do have its pricing problem. Every AI product eventually meets a user who costs more than they pay.
Make the limit legible. Show usage where the user works, and say what happens at zero before they get there. Then make the overflow path obvious: a credit pack, a higher tier, or a reset time.
Do not copy the scale. A solo founder cannot verify every account by hand, but a new account can start lower and earn more with a payment history. The guide to pricing an AI product covers packaging, and the credits explainer covers the unit. If cost is the real problem, start with the LLM cost guide.
The full argument, with a step-by-step plan, is in the guide to setting usage limits for an AI product.
Where ZeroRaised fits#
ZeroRaised is a launch and discovery site for products built without institutional venture capital or accelerator equity. If you sell an AI product, submit it here and put your pricing and limits on the page, because a clear ceiling is a selling point. The newsletter carries new analysis like this. We make no promise of traffic or results.
Sources#
Frequently asked questions
What changed about Claude Max and Ultra plans?
Anthropic renamed the $100 Max 5x plan to Max and the $200 Max 20x plan to Ultra, kept the prices the same and announced it on X on 30 October. Verified Ultra accounts no longer have a limit that resets every five hours, while weekly limits stay in place.
Does Claude still have a five-hour limit?
Most accounts still have a limit that resets on a rolling five-hour session window, and paid plans add weekly limits on top. The exception is verified Ultra accounts, which no longer have that five-hour reset. Newer accounts keep a temporary limit until Anthropic verifies them.
Why did OpenAI add card and device limits?
OpenAI describes card and device limits as part of ChatGPT's privacy and security protections: too many subscriptions on one payment card or device can block a new purchase. The rules apply to ChatGPT Go, Plus, Pro 100, Pro 200 and Pro 500, and buying credits or resetting usage will not clear the restriction.
What does it mean that ChatGPT and Codex share an allowance?
From December, the usage allowance on Plus, Pro and Business plans will be shared across ChatGPT and Codex, measured in tokens. Advanced models and complex tasks use more allowance, while instant chat does not draw from it. At the limit a user can wait, upgrade or buy credits.
What can a small AI product copy from this?
Name tiers by position rather than by a quota, pool usage across surfaces, sell overflow credits instead of only blocking, verify new accounts instead of pricing everyone out, and explain any change to the numbers clearly.