Our AI Transparency Policy: Humans Write Our Content

TL;DR- Key Takeaways of Our Policy

AI is a Tool and Can be Used for Rote Tasks, but Not for Writing Content or Presented as Human-Written Content.

This is a rule we are very strict with. For the content created by, published on, or distributed through Howdy Curiosity, we expressly forbid the use of churning out AI-generated content in bulk.

To ensure that our content meets our readers' expectations and that we're being transparent in all things, we have defined use cases for when AI tools are appropriate to use and drawn hard lines around when they are not acceptable to use. 

These rules apply to any content we create in-house, or third-party submissions to our blog or community.

WHERE IS AI ALLOWED?

Acceptable AI Use Cases

We allow and may use AI tools such as ChatGPT in the following use cases: 

  1. Suggesting recommendations for making headlines more concise or compelling (e.g., prompt: recommend 5 alternative headlines for the following blog excerpt. The current headline is XYZ, but I want something more engaging for readers...)
  2. Describe and generate possible featured images or in-content images. (e.g., prompts: create a description of an image that conveys ABC; generate an image based on your description)
  3. Validate brand voice (e.g., prompt is a custom GPT trained on our brand voice: create a bulleted list of editorial recommendations to make the following text better align with our brand voice)
  4. Search Recommendations: Our bookshop (i.e., the shop.howdycuriosity.com domain) uses AI in the search bar to help customers find the title they're looking for when they don't query an exact match, and it makes recommendations for similar titles a user may like.

WHERE IS AI BANNED INTERNALLY?

Prohibited AI Use Cases

We strictly forbid the use of AI tools in the following ways:

  1. Writing blog posts or other types of long-form content. Everything we publish must be written by a human and capture their unique point of view in a way that adds value to the topic for our readers. 
  2. Creating emails to do outreach or marketing. 
  3. Assuming the role or presentation of a human. If we say a member of our team is doing something– such as providing book recommendations– then it is a human member. We will not pass off AI activities as human activities. 
  4. Generating fake contributing authors, endorsements, reviews, followers, or customers. Using AI to create fake profiles in any capacity- including marketing or sales purposes– is not allowed.