Apple gives nod to ChatGPT-driven app amid concerns

Apple Inc. AAPL green up pointing triangle has approved an email-app update after initially scrutinizing whether a feature in the software that uses language tools powered by artificial intelligence could generate inappropriate content for children.

The app, BlueMail, was approved following assurances from its developer that it features content moderation, according to Ben Volach, co-founder of the app-maker, Blix Inc.

The Wall Street Journal reported on Thursday that the update, which included a new feature powered by language chatbot ChatGPT, was held up due to Apple’s request that the app add content moderation or be restricted to ages 17 and older. The app was previously available for ages 4 and older.

Blix told Apple its update includes content moderation and suggested that the company should make public any new policies about the use of ChatGPT or other similar AI systems in apps. The BlueMail update was approved without changes on Thursday evening. The app is still available for users aged 4 and older.

BlueMail’s new feature uses OpenAI’s ChatGPT, an artificial-intelligence system capable of answering questions or writing short essays, to help automate the writing of emails using the contents of prior emails and calendar events.

The news of Apple’s initial rejection of BlueMail’s ChatGPT feature highlighted the growing concerns around new uses of language-generating AI tools. ChatGPT allows users to converse with an AI that appears humanlike, but early testing has shown the AI producing incorrect information as well as strange and sometimes hostile responses.

Although Apple took action on the AI feature, there have been no additional formal App Store guidelines from Apple regarding the use of so-called generative AI in apps.

Apple has long said that it has to curate and review each app that goes through its App Store in order to keep its products safe for users. Wall Street Journal