Notifying users of offensive content
Techniques are described that determine when content to be shared by a user may be offensive, and providing the user with a notification that the content may be offensive. In some examples, the user may be provided with a notification that includes a selectable option allowing the user to withdraw or undo the content from being shared, additional information regarding why the content was determined to be offensive, content sharing guidelines, and/or advice for how to revise the content to be less offensive. In some examples, the notification may be presented for a predetermined period of time, and the content may be held in a pending state and not shared with other users until expiration of the predetermined period.
1. A method comprising:
sending, to a computing device associated with a first user account, content originating from a second user account;
receiving, from the computing device associated with the first user account, a comment associated with the content;
inputting the comment into a machine-learned model trained to identify potentially offensive comments;
receiving, from the machine-learned model, a score associated with an offensiveness of the comment;
comparing the score to a threshold score corresponding to an offensiveness level;
sending, to the computing device associated with the first user account and based at least in part on the score being greater than the threshold score, an instruction to present a notification associated with the comment for a duration of time;
withholding the comment from being transmitted to the second user account or one or more other user accounts for the duration of time; and
transmitting the comment to at least one of the second user account or the one or more other user accounts responsive to the duration of time elapsing.
2. The method of claim 1 , wherein the instruction to present the notification further instructs the computing device associated with the first user account to display a visual indicator to indicate the duration of time.
3. The method of claim 1 , wherein the instruction is a first instruction and the duration of time is a first duration of time, the method further comprising sending, to the computing device associated with the first user account, a second instruction to display the comment in a pending state for a second duration of time.
4. The method of claim 1 , wherein the instruction is a first instruction, the method further comprising sending, to the computing device associated with the first user account, a second instruction to present a control usable to undo the comment.
5. The method of claim 4 , further comprising:
receiving, from the computing device associated with the first user account, an indication of a selection of the control to undo the comment; and
sending, to the computing device associated with the first user account, a third instruction to display the comment in an editable field that enables the user to edit the comment.
6. The method of claim 1 , method further comprising:
receiving, from the computing device associated with the first user account, a request for information about the offensiveness of the comment; and
transmitting, to the computing device associated with the first user account, the information.
7. The method of claim 1 , wherein the content comprises at least a portion of a profile post, a story, or a direct message.
8. The method of claim 1 , wherein the threshold score corresponding to an offensive level is based at least in part on the comment including at least one of bullying, hate speech, taunting, threats, blackmail, harassment, or promotion of self-harm.
9. The method of claim 1 , wherein the comment includes an indication of the second user.
10. A system comprising:
one or more processors; and
computer-readable media storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:
sending, to a computing device associated with a first user account, content originating from a second user account;
receiving, from the computing device associated with the first user account, a comment associated with the content;
inputting the comment into a machine-learned model trained to identify potentially offensive comments;
receiving, from the machine-learned model, a score associated with an offensiveness of the comment;
comparing the score to a threshold score corresponding to an offensiveness level; and
sending, to the computing device associated with the first user account and based at least in part on the score being greater than the threshold score, an instruction to present a notification associated with the comment for a duration of time and to display a visual indicator to indicate the duration of time.
11. The system of claim 10 , the operations further comprising transmitting the comment to at least one of the second user account or one or more other user accounts responsive to the duration of time elapsing.
12. The system of claim 10 , wherein the instruction is a first instruction and the duration of time is a first duration of time, the operations further comprising sending, to the first computing device associated with the first user account, a second instruction to display the comment in a pending state for a second duration of time.
13. The system of claim 10 , wherein the instruction is a first instruction, the operations further comprising sending, to the first computing device associated with the first user account, a second instruction to present a control usable to undo the comment.
14. The system of claim 13 , the operations further comprising:
receiving, from the computing device associated with the first user account, an indication of a selection of the control to undo the comment; and
sending, to the computing device associated with the first user account, a third instruction to display the comment in an editable field that enables the user to edit the comment.
15. The system of claim 10 , wherein the content comprises at least a portion of a profile post, a story, or a direct message.
16. One or more computer-readable media storing instructions that, when executed by one or more processors of a server computing device, cause the server computing device to perform operations comprising:
sending, to a computing device associated with a first user account, content originating from a second user account;
receiving, from the computing device associated with the first user account, a comment associated with the content;
inputting the comment into a machine-learned model trained to identify potentially offensive comments;
receiving, from the machine-learned model, a score associated with an offensiveness of the comment;
comparing the score to a first threshold score corresponding to a first offensiveness level;
preventing the content from being sent to the computing device associated with the first user based at least in part on the score being greater than the first threshold score;
comparing the score to a second threshold score corresponding to a second offensiveness level; and
sending, to the computing device associated with the first user account, an instruction to present a notification associated with the comment based at least in part on the score being less than the first threshold score and greater than the second threshold score.
17. The one or more computer-readable media of claim 16 ,
wherein the notification includes a control usable to undo sharing the content based at least in part on the score being less than the first threshold score and greater than the second threshold score.
18. The one or more computer-readable media of claim 16 , wherein the instruction to present the notification includes a duration of time to present the notification.
19. The one or more computer-readable media of claim 18 , wherein the instruction to present the notification further instructs the computing device associated with the first user account to display a visual indicator to indicate the duration of time.
20. The one or more computer-readable media of claim 18 , the operations further comprising transmitting the comment to at least one of the second user account or one or more other user accounts responsive to the duration of time elapsing.