Humans have given it AI blog.

Anthropic quietly launched Claude a week ago, explaining that it was a new page on its website, generated mainly by Claude, the company’s family of AI models. The blog covers posts on technical topics related to various Claude use cases, such as “Simplifying complex code bases with Claude”, which aims to demonstrate Claude’s writing skills.

It is not clear how many posts Claude's original writing introduces in Claude. According to a spokesperson, the blog was overseen by anthropomorphic “subject experts and editorial teams” who “enhanced” Claude’s draft which included “instances, practical examples and (…) contextual knowledge.”

"It's not just the output of Vanilla Claude - the editing process requires human expertise and iterating," a spokesperson said. "From a technical point of view, Claude explains that Claude (creates) educational content, along with our team comments, refines and enhances a collaborative approach."

There is no obvious view in the home page explained by Claude, which describes it: "Welcome to the small corner of the human universe, where Claude writes on every topic in the sun." It may be easily misled to think that Claude is responsible for the end-to-end copy of the blog.

Anthropic said it believes Claude interprets it as "demonstrating how human expertise and AI capabilities work together", starting with educational resources.

"Claude explained early examples of how teams can use AI to enhance their work and provide users with greater value," the spokesperson said. "Instead of replacing human expertise, we show how AI can scale up topics that subject experts can accomplish (…) We plan to cover topics from creative writing to data analytics to business strategy."

Anthropic experimented with AI-generated copy, which was far from the first to be elucidated a few months after Rival Openai said it had developed a model for creative writing. Meta's Mark Zuckerberg said he wanted to develop an end-to-end AI advertising tool, and OpenAI CEO Sam Altman recently predicted that AI could one day handle "95% of marketers using the work of institutions, strategists and creative professionals today."

Elsewhere, publishers have also piloted AI news tools to increase productivity and, in some cases, reduce hiring demand. Gannett is particularly aggressive, launching an AI-generated sports review and a summary below the headlines. Bloomberg added an AI-generated summary at the top of its April post. Last week, 21% of employees were abandoned by business insiders, who pushed writers to switch to assisted AI tools.

Even the heritage channels are investing in AI, or at least suggest possible ambiguity. The New York Times reportedly encouraged employees to use AI advice to raise editors, headlines and even questions in interviews, while the Washington Post is said to be developing an "AI-powered Story Editing" called Ember.

However, many of these efforts are not going well, mainly because today's AI can easily make up for this goal confidently. According to Semafor, business insiders were forced to apologize to employees after recommending books that did not appear to exist but could be produced by AI. Bloomberg must correct dozens of AI-generated article summary. G/O Media's erroneous flow AI writing feature, opposing the desire to edit, has caused widespread ridicule.

The personification spokesman noted that despite the company being immersed in AI-driven blogging, the company is still hiring across markets, content and editorials, and “many other areas involving writing.” Put it into practice.