Anthropic Is Raising Prices And Pissing People Off
Anthropic is changing its Claude subscription model starting June 15th, shifting third-party app usage to a credit-based system billed at API rates, which users argue significantly reduces value. Despite user frustration, Anthropic has surpassed OpenAI in business adoption as of April.
Summary
Anthropic made two notable announcements this week. First, they increased weekly limits for Claude Code by 50% through July 13th, which was generally positive news. However, a second announcement generated significant backlash: starting June 15th, Anthropic is changing how Claude subscriptions work for usage outside of Claude Code.
Under the new model, users accessing Claude through third-party applications like Open Claw or Hermes — tools built on the agent SDK — will receive a monthly credit allocation based on their subscription tier. Once those credits are exhausted, usage will be billed at standard API rates. Critics in the community are framing this as a major reduction in value rather than a new feature, arguing that API billing rates are expensive enough that even $100–$200 in credits could burn through in just a few hours of heavy use, rendering the subscription impractical for serious development work.
Despite this controversy, Anthropic appears to be gaining ground in the enterprise market. Business adoption of Anthropic rose 3.8% in April, bringing its share to 34.4%, while OpenAI's adoption fell 2.9% to 32.3% — marking a notable shift in competitive positioning between the two AI companies.
Key Insights
- Anthropic increased Claude Code weekly limits by 50% through July 13th, but the positive news was overshadowed by a concurrent pricing change announcement.
- Starting June 15th, Claude users accessing the platform via third-party apps like Open Claw or Hermes will be given a monthly credit allocation, after which they are billed at API rates rather than enjoying flat subscription access.
- Community members are characterizing the new credit system as 'a massive nerf, not a new feature,' arguing that API-rate billing makes even $100–$200 in credits burn out in just a few hours of heavy use.
- The pricing change is seen as particularly damaging for developers doing serious work through third-party agent SDK-based applications, where high usage volume makes API-rate billing prohibitively expensive.
- Despite user frustration over the pricing changes, Anthropic surpassed OpenAI in business adoption in April, with Anthropic reaching 34.4% (+3.8%) versus OpenAI's 32.3% (-2.9%).
Topics
Transcript
[0:00] Anthropic announced that they were increasing the weekly limits for Claude Code by 50% through July 13th. However, a lot of people aren't as excited as you'd think about this because it was also announced this week that starting June 15th, they're changing the way Claude subscriptions work. If you want to use Claude outside of Claude Code, you know, through things like Open Claw or Hermes or those kinds of things, well, they're going to give you a certain amount of credits every month depending on the plan that you have. And then once your credits run out, you'll [0:31] then start getting build at the API rate. The thread that follows essentially says, "This is a massive…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Matt Wolfe
The ONLY AI Benchmark You Need!
A developer created "Buccy Bench," a humorous yet functional AI benchmark that tasks different language models with drawing Gary Busey as SVG code rather than images. The benchmark tracks model evolution over time while measuring performance metrics like cost, tokens, and execution speed.
GLM-5.2 - The Open Model That's As Good As Opus!
A comprehensive review of GLM-5.2, an open-weight Chinese AI model with a 1 million token context window, demonstrating its capabilities for coding, document analysis, and agentic workflows at significantly lower costs than frontier models like Claude Opus and GPT-4.5. The speaker tests various use cases including website building, Chrome extensions, game development, and data organization, concluding it's valuable for long, code-heavy, token-expensive tasks despite not universally outperforming closed-source alternatives.
Don't Fall For This AI Trap
The speaker emphasizes that power users distinguish themselves by knowing what NOT to automate with AI, rather than automating everything. They argue that AI works best for clear, straightforward tasks but struggles with nuanced, artistic work requiring consistency—using their failed YouTube thumbnail automation as an example.
AI News: Fable Banned, New Open-Source Leader, Midjourney Shocker
This AI news roundup covers the US government forcing Anthropic to shut down its Fable 5 and Mythos 5 models due to a security vulnerability jailbreak, the release of a competitive open-source model GLM 5.2 by ZAI, and MidJourney's surprising pivot into medical imaging technology with a new ultrasound-based body scanner.
AI News: Claude's Massive Leap & Siri Gets Good!?
This AI news roundup covers the release of Claude Fable 5 (a Mythos-tier model from Anthropic) and its controversial safety restrictions, Apple's WWDC AI announcements including a major Siri overhaul, and updates from Google including NotebookLM upgrades and real-time translation. Additional rapid-fire items include OpenAI and SpaceX IPO filings, ChatGPT email sending, and a teased Midjourney hardware device.