On Monday, OpenAI disclosed a collaboration with the nonprofit Common Sense Media to develop AI guidelines and educational resources tailored for parents, educators, and adolescents. This initiative encompasses the curation of family-friendly GPTs within OpenAI’s GPT store, with the primary goal of addressing concerns regarding the impact of AI on young individuals.
Common Sense Media, renowned for its evaluations of movies and TV programs intended for children, has recently expanded its focus to include AI, incorporating reviews of AI assistants on its platform.
Stressing the significance of educating children on responsible AI usage, Common Sense Media stated, “AI isn’t going away, so it’s crucial that we empower youngsters to utilize it conscientiously.” This sentiment was echoed by OpenAI CEO Sam Altman and Common Sense Media CEO James Steyer during the announcement at the Common Sense Summit for America’s Kids and Families in San Francisco, a well-publicized event on the social media platform X.
Altman expressed enthusiasm about the benefits of AI for families and teenagers, emphasizing that the partnership with Common Sense will enhance safety measures, promoting the confident use of their tools by families and teens.
Although the official news release may seem somewhat vague, Steyer affirmed that their collaborative efforts will focus on educating families and educators about the safe and responsible utilization of ChatGPT to mitigate any inadvertent repercussions of this advancing technology.
The partnership primarily aims to imbue OpenAI’s GPT store with a family-friendly aura, with Common Sense Media contributing to the curation of GPTs aligned with their ratings and standards.
In a separate context, Common Sense Media began evaluating AI assistants on its platform last year, potentially introducing conflicts of interest in light of the new partnership with OpenAI. However, there is no apparent bias towards OpenAI evident thus far.
For instance, Common Sense Media’s review of ChatGPT characterizes the AI assistant as “A potent, occasionally risky chatbot suitable for individuals aged 13 and above, ideal for creative use rather than factual queries.” Notably, ChatGPT received a three-star rating and a 48 percent privacy score in the review, last updated on October 13, 2023.
Comparatively, Google Bard garnered a three-star overall rating and a 75 percent privacy rating, while Stable Diffusion, an image synthesis model, received a one-star rating due to safety concerns. OpenAI’s DALL-E secured two stars and a 48 percent privacy rating.
Common Sense Media’s evaluations of each AI model are comprehensive and factually sound, even with the partnership with OpenAI. Despite the modest ratings assigned, it is evident that most AI models are still in the nascent stages of development, indicating the potential for future advancements in generative AI.
