A rule barring such hate speech targeting religious groups now applies to "language that dehumanises on the basis of age, disability, or disease," the Twitter safety team said in an online post.
"Our primary focus is on addressing the risks of offline harm, and research shows that dehumanising language increases that risk," it said.
Twitter will delete tweets already posted that break the rules, relying on users to report offending comments.
Examples of rule-breaking tweets included posts that refer to people with a disease as "rats that contaminate everyone around them" or people with a particular disability as "subhuman."
Twitter and other social networks have been struggling with removing hateful and abusive content while remaining open platforms.
The online services have also faced accusations that the curbs suppress some political views.