Anthropic adjusts Claude session limits and says users will hit their limits faster during peak hours, amid compute strain due to Claude's new popularity
here's what changedMarcus Schuler /Implicator.ai:Anthropic Knew the Math. It Sold the Tickets Anyway.Ben Patterson /PCWorld:Anthropic confirms it's been ‘adjusting’ Claude usage limits
Business InsiderBrent D. Griffiths
Context & Ripple Effects
Anthropic’s latest warning is part of a recurring pattern in Claude access management. In 2025, it said new Pro and Max rate limits would chiefly target users running Code continuously in the background, following reports that Max subscribers faced restrictive Claude Code limits.
The significance is not merely a revised quota: peak-hour limits make available access depend on when customers use Claude. That sharpens the practical distinction between nominal plan access and capacity actually deliverable under load.
First-order effects
Claude users will exhaust their session allowances more quickly at peak times, reducing uninterrupted use for chat and coding workloads during those periods.
Anthropic can ration constrained compute immediately through usage policy rather than broadly removing access; customers whose work is time-sensitive bear the operational impact.
Second-order effects
Heavy users may shift work to off-peak periods, shorten workflows, or seek alternative tools when availability matters more than a plan’s headline limits.
The change raises the stakes for Anthropic’s plan design and customer communication: its earlier planned Pro and Max rate limits already identified sustained background Code use as a pressure point.
Third-order effects
If peak-sensitive limits persist, AI subscriptions may increasingly be defined by managed capacity and service conditions, not simply a fixed set of model features.
Recurring access restrictions could make transparent inference budgeting and differentiated service tiers a more important competitive dimension, though this report alone does not establish a durable pricing-model change.
The trend: Frontier AI providers are moving toward dynamic capacity allocation as demand for compute-intensive usage outpaces the predictability of flat subscription access.
To manage growing demand for Claude we're adjusting our 5 hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged. During weekdays between 5am-11am PT / 1pm-7pm GMT, you'll move through your 5-hour session limits faster than before.
I think people are overreacting to this. It sucks, but they clearly have a real compute shortage and this was the best path they had without cutting your usage massively
We've landed a lot of efficiency wins to offset this, but ~7% of users will hit session limits they wouldn't have before, particularly for pro tiers. If you run token-intensive background jobs, shifting them to off-peak hours will stretch your session limits further.
Overall weekly limits stay the same, just how they're distributed across the week is changing. I know this was frustrating. We're continuing to invest in scaling efficiently. I'll keep you posted on progress.
You're going from rewarding users using it during off-peak hours (double usage which ends March 28th) to punishing them for using during peak hours. I think that's a terrible move if you want happy customers.
meanwhile, you can enjoy Codex with 2x rate limits for the next week 24x7 across all ChatGPT subscriptions! npm i -g @openai/codex time to build is NOW!!
Thariq: “I need to give our customers less for the same price. How can I obfuscate that as much as possible” Claude: “How about 'You'll move through your 5-hour session limits faster than before'”
You make this announcement BEFORE you make the change. Good riddance... what a shame @AnthropicAI and @claudeai. Customer service isn't hard, @DarioAmodei. This was a rug pool and y'all got caught. I hope the @FTC fines you.
Devs who can code also WITHOUT AI as well looking to became 10x more valuable They are the ones who won't panic or be idle when their Claude quota runs out... So much for all the advice on how learning to code is not worth it any more...
My counterpoint to Citrini Research's prediction of mass unemployment by 2028 due to AI agents is we don't have the data center capacity. — Anthropic is tightening usage caps and expects 7% of users to hit peak-hour limits. Token cost and compute are still real bottlenecks to …