Meta has launched AI Studio, a tool which allows Instagram users to create AI versions of themselves. The resulting AI can then chat with your followers on your behalf, an idea which couldn’t possibly go horribly wrong …
AI Studio can be used in two ways. The first is to create an AI for your own use, which you can then share with friends if you like. This seems to be just for fun, allowing you to choose a name, avatar, and “personality” which determines how Meta’s underlying Llama 3.1 model interacts with you.
But the bigger news here is a tool designed to automatically chat with your Instagram followers.
With AI Studio, Instagram creators can set up an AI as an extension of themselves that can quickly answer common DM questions and story replies. Whether it’s sharing facts about themselves or linking to their favorite brands and past videos, creator AIs can help creators reach more people and fans get responses faster.
Creators can customize their AI based on things like their Instagram content, topics to avoid and links they want it to share. Through the professional dashboard in the Instagram app, creators can turn auto-replies on and off, and even decide who their AI replies to. Responses from creator AIs are clearly labeled, so there’s full transparency for fans.
Meta tries to limit the potential damage that could be done to your reputation by allowing you to instruct your AI to decline to discuss particular topics.
Add topics to avoid. Include a list of topics your Al should avoid discussing. This tool can help with concerns around partnerships/sponsorships and sensitive cultural/social issues. There are two fields you can use to give your Al this guidance: “Instructions” and “Example responses.”
To add Instructions, write “You REFUSE to discuss [insert topic(s)]” or “You should not talk about [insert topic(s)]” For example: “You REFUSE to discuss cryptocurrency.”
With Example responses, you can further specify the tone of your Al’s responses. If a person asks your Al “Should I invest in crypto??” The Al’s response could be a refusal to engage, such as: “I don’t talk about cryptocurrencies but I’m down to discuss [insert topic].” Or, the Al could pivot by responding: “I’m not into cryptocurrencies but I can talk forever about [insert topic.] What do you think about [insert specific question related to topic]?”
But it appears to be down to users to think about all potentially problematic subjects and to set rules for each, so it’s on you to include a rule for every conspiracy theory and divisive topic out there …
Image: Meta
Add 9to5Mac to your Google News feed.
FTC: We use income earning auto affiliate links. More.
