Anthropic is facing major backlash from developers and users who report a decline in the performance of its Claude AI models. Users say the system is increasingly failing to follow instructions, producing more errors, and struggling with complex multi-step workflows.
Claim 30% Off TipRanks
New trading tool for NVDA bearsUsers Flag Claude AI Issues
Claude AI performance complaints have spread across developer communities such as GitHub, X (formerly Twitter), and Reddit (RDDT) in recent weeks. Users report weaker results in sustained reasoning tasks, cases where the AI model abandons tasks midway and generates outputs that are hallucinations or contradictory.
Some developers also point to weaker coding results from Claude Code and a shift toward faster but less detailed responses. Others say that Claude Opus 4.6 struggles with longer, more complex prompts. These issues have led to claims of “AI shrinkflation,” a term used to describe reduced output quality without clear changes in pricing or usage limits.
Users Accuse Anthropic of Reducing Claude AI Capability
Several developers and users have accused Anthropic of weakening its Claude AI models, pointing to recent system changes and tighter usage controls. While responding to a user complaint, Boris Cherny, who leads Claude Code, said the firm lowered the default “effort” level to “medium” following feedback that the model was using too many tokens per task.
However, some users said this change was not clearly communicated. Anthropic has also experienced several outages and introduced stricter usage limits during peak periods, which added to criticism. Attention has now turned to compute capacity. A leaked memo from OpenAI said Anthropic made a “strategic misstep” by not securing enough compute and was “operating on a smaller curve” than competitors.
The private company recently expanded deals with Google (GOOGL) and Broadcom (AVGO) to secure about 5 gigawatts of TPU and chip supply. While no direct link has been confirmed, some users cite computing constraints as a possible factor affecting Claude’s performance.
Are AI Shares a Good Investment?
AI-related stocks have delivered strong returns for investors over the past few years, especially shares of U.S. public companies. Investors interested in exploring these stocks can watch for well-known players such as Nvidia (NVDA), Micron (MU), Microsoft (MSFT), Meta Platforms (META), and GOOGL. These stocks have been rated a “Strong Buy” by Wall Street analysts tracked on TipRanks. For more information on their performance, ratings, and price targets, visit the TipRanks Stocks Comparison Center.
