Every Anthropic announcement follows a recognizable structure: a paragraph on safety, a reference to responsible AI development, a mention of their Constitutional AI approach (a method of training models by having them evaluate responses against a set of ethical principles), and then - somewhere in the middle - the actual news.
It's become predictable enough that users can practically fill in the template before reading. New model launch? Safety first. New research paper? Safety first. New enterprise deal? Safety first.
This isn't an accident. Anthropic was founded in 2021 by former OpenAI researchers who left partly over safety concerns, and responsible development isn't just marketing - it's the company's stated organizing principle. Their Responsible Scaling Policy, published in 2023 and updated since, commits them to specific safety evaluations before deploying more powerful models.
The criticism isn't that they care about safety. It's that every communication leads with the same framing regardless of context, which makes the safety language feel increasingly like boilerplate. When every announcement is a safety story, none of them are.
For users who just want to know whether to upgrade their plan or try a new feature, Anthropic's communications style adds friction. OpenAI ships a model and leads with what it can do. Anthropic ships a model and leads with how carefully they approached it. Users have to read further to get to the capabilities.
Neither approach is objectively better - they reflect genuinely different company priorities. But there's a practical cost to Anthropic's formula: it consistently buries the product information that most users came to find.