The policy changes were specifically aimed at protecting women who unknowingly or unwillingly had nude pictures of themselves distributed online or were subject to unwanted sexual advances. They would also aim to shield groups subject to hateful imagery, symbols and threats of violence.
In an email Twitter shared with The Associated Press Tuesday, Twitter's head of safety policy outlined the new guidelines to the company's Trust and Safety Council, a group of outside organizations that advises the company on its policies against abuse.
The company said it would enact the changes in the weeks ahead. News of the policy changes was first reported by Wired.
Among the changes, Twitter said it would immediately and permanently suspend any account it identifies as being the original poster of "non-consensual nudity,'' including so-called "creep shots'' of a sexual nature taken surreptitiously. Previously, the company treated the original poster of the content the same as those who re-tweeted it, and it resulted only in a temporary suspension.
It said it would also develop a system allowing bystanders to report unwanted exchanges of sexually charged content, whereas in the past it relied on one of the parties involved in the conversation to come forward before taking action.
Twitter also said it would take new action on hate symbols and imagery and "take enforcement action against organizations that use/have historically used violence as a means to advance their cause,'' though it said more details were to come.
While it already takes action against direct threats of violence, the company said it would also act against tweets that glorify or condone violence.
On Friday, Dorsey foreshadowed the coming policy changes in a series of tweets, saying the company's efforts over the last two years were inadequate.