Coding Tools
Anthropic blocks Claude Code from cloning existing games
The Verge AIhigh confidence
Why it matters
- →Anthropic actively restricts Claude Code from generating code for direct clones of existing games.
- →The update on Opus 5.5 distinguishes between original creation and IP infringement in coding assistance.
- →Developers should expect stricter refusals when attempting to replicate protected game mechanics or assets.
Anthropic has updated the system prompt for Claude Code on the Opus 5.5 model to prevent the AI from generating code that recreates specific existing video games. Addy Osmani, Anthropic’s VP of Product, confirmed the change aims to stop IP infringement while allowing users to build original games or titles within a genre using their own designs. The update specifically targets direct clones and decompilations, distinguishing them from original creative work. This policy adjustment reflects growing concerns over AI-generated content violating intellectual property rights in the gaming sector.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Anthropic to Watermark AI-Generated Text
TechCrunch AI · August 11, 2026 · Related
- 2Anthropic to Add Watermarks to AI Content
The Verge AI · August 11, 2026 · Related
- 3Anthropic adds invisible watermarks to Claude outputs
The Rundown AI · August 12, 2026 · Related
- 4Anthropic embeds invisible watermarks in Claude outputs
The Rundown AI · August 12, 2026 · Related
- 5Anthropic's Claude Adds Watermarks to AI Outputs
TechCrunch AI · August 12, 2026 · Related
- 6Developers Bypass Claude's Invisible Watermarks
WIRED AI · August 19, 2026 · Same story
- 7Anthropic releases Claude Opus 5.5 with tighter security
The Verge AI · September 22, 2026 · Related
- 8Anthropic releases official Claude Code playbook for Opus 5.5
Duncan Rogoff · September 23, 2026 · Same story
More from The Verge AI
© The Verge AIMarket & Regulationbusiness
OpenAI pilots grid protection program for data centers
OpenAI is flipping the script on the growing backlash against AI infrastructure by launching a pilot in Georgia where its data center acts as a backup power source for up to 700,000 homes during emergencies. This 'Grid Protectors Program' reframes massive energy consumers as critical grid stabilizers, directly addressing local opposition fueled by skyrocketing electricity demands and environmental concerns. By offering flexibility to reduce demand during peak strain while providing steady baseline load, OpenAI aims to turn a liability into a community asset. It is a strategic PR move that acknowledges the physical reality of AI's energy hunger without solving the underlying resource scarcity.
The Verge AI
© The Verge AIMarket & Regulationother
NYT sues Brave for AI scraping and paywall bypass
The New York Times has escalated the copyright war by suing Brave Software, alleging it illegally scraped content to train AI models and sold access to that data. The complaint names Anthropic, Perplexity, Mistral, xAI, and Cohere as recipients of this 'stolen' index, framing Brave as a middleman laundering pirated journalism for generative AI giants. This moves the legal battle from abstract fair use debates into specific allegations of unauthorized commercial distribution and paywall circumvention. It signals that major publishers are now targeting the infrastructure layer of AI data collection, not just the model builders.
The Verge AI
© The Verge AICoding Toolscoding
DIY Muse Gadget proves Meta's hardware is just a dev kit
Meta’s Muse Charm isn’t a finished product; it’s an invitation to hack. A non-coder used Claude Code to transform a $50 Waveshare ESP32-S3 dev board into a fully functional AI companion, adding ElevenLabs voice synthesis, custom Tamagotchi-style minigames, and remote camera controls that Meta’s official firmware lacks. This reveals a critical truth: the value of Meta’s new hardware ecosystem lies not in its out-of-the-box utility, but in its open-source API allowing anyone to rewrite the experience. The gap between Meta’s polished demo and the actual user reality is bridged entirely by third-party AI coding tools.
The Verge AI
More in Coding Tools
Coding Toolscoding
Claude Code v2.1.317 patches security and stability
This release prioritizes trust and reliability over new features, addressing critical friction points for enterprise users. The most significant addition is a safety check that prompts before sending account emails to external networks, directly tackling data leakage concerns in automated workflows. For developers managing scale, the new gateway admin API cost report provides granular visibility into spend per developer and IdP group, solving a major blind spot in self-hosted deployments. Stability fixes for session persistence, proxy timeouts, and credential handling ensure that long-running agent tasks don't silently fail or leak data.
Claude Code Releases
Coding Toolscoding
llama.cpp b12053 updates Apple memory and adds ROCm 10
This release quietly fixes a memory allocation quirk on Apple Silicon by switching to vm_allocate, which should stabilize buffer handling across macOS and iOS. More notably, it brings ROCm 10.0 support to both Linux and Windows, finally giving AMD GPU users parity with CUDA builds in the standard distribution. The update also extends CUDA 13 compatibility to ARM64 architectures, broadening the hardware base for high-performance inference. It’s a maintenance-heavy release that solidifies llama.cpp as a cross-platform runtime rather than just a local Mac tool.
llama.cpp Releases
Coding Toolscoding
llama.cpp b12057 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing first-class ROCm 10.0 support to both Linux and Windows. AMD GPU users no longer need workarounds, effectively closing a long-standing parity gap with NVIDIA's CUDA ecosystem. Equally notable is the new Snapdragon build for Linux, which unlocks local AI on ARM-based laptops using Adreno GPUs and Hexagon NPUs. While KleidiAI on Apple Silicon is temporarily disabled, the expansion into AMD and Qualcomm hardware signals a strategic shift toward broadening accessible local inference beyond just NVIDIA.
llama.cpp Releases