Gemini is reportedly getting a content output filter soon, and it could make interacting with it a more enjoyable experience for a wider range of users.
The way Gemini works right now, you can use it for tons of different things. You can ask it to set timers, ask it questions, and even task it with locating restaurants for you. Its Deep Research also now supports more languages. These are just the tip of the iceberg. What it can’t currently do is filter out the responses for you that were user-designated. But that seems to be changing. As noted by Android Authority, the latest version of the Google app beta is preparing to add a new option for Gemini in the settings. One that’s called “Content Filters,” and does exactly what it says.
It lets you filter out responses that you don’t want to receive. Gemini already sort of does this on its own. However, it’s not a user content filter. It’s more of a set of broad categories Gemini is coded to try and avoid. Such as hate speech or sexual harassment. This new set of filters is user-controlled. So what you can filter out from the responses will presumably be much more granular.
Gemini Content Filter options aren’t user-facing yet
As of right now the feature isn’t live but Android Authority was able to activate the menu to get it to show up. It doesn’t work, however. So it’s still unclear what it will be able to do in its entirety and how fine-tuned the user can make these content filters.
As Android Authority notes, it’s entirely possible that Google may let users only filter content on a broader scale. Which is basically what it does on its own right now for specific things like those mentioned above. Or Google could allow users to filter out very specific things. Not too unlike the way you can block words in X to avoid seeing any posts that mention those exact words or phrases.
There’s no telling when the feature will be live
Since Google hasn’t announced this feature at all, and it’s not currently live or publicly available, it might still be a while before it rolls out to users. It’s also worth remembering that even in the most recent beta, getting this option in settings to show up had to be activated and wasn’t already there like it would be if Google was ready for the feature to be tested.
On top of that, tapping on the menu doesn’t currently do anything. At least, not other than take you to an inactive URL on the Gemini page. It seems clear though what Google probably wants this feature to do. As it’s reasonable not to want the AI tool you’re pushing pretty heavily to respond to users with potentially offensive stuff. In addition to this new feature, Google also recently announced Gemini 2.0.
2024-12-27 15:07:50