For about an hour on a recent Thursday morning, the verified Twitter account of the late New York Times media columnist David Carr belonged to an apparent spambot. The account’s name changed to “Miranda Davis,” and “Miranda” tweeted, “I love role-playing games and sex.” The tweet shocked many of Carr’s hundreds of thousands of followers: Carr died in 2015.

The spambot wasn’t the first ghost. Before the Carr account began to tweet, his old email address was apparently sending spam to his contacts. “There is almost nothing worse than getting spam from David Carr’s old/hacked email address,” Clara Jeffery, the editor-in-chief of Mother Jones tweeted. Carr’s Twitter account was restored within hours.

It might be kind of unusual to witness a high-profile Twitter account like Carr’s fall victim to a spamming sexbot, but the unintended digital afterlives of people who have died appear daily. Some are experienced in relative isolation: the email of a dead friend appearing while searching a contact list, for instance. When the afterlives continue on social media, however, the experience is public, and more collective.

At this point, social networks have some ways to handle these afterlives, for the benefit of the living who knew them. Facebook and Instagram allow family members to request account deletion for a deceased person, and for the memorialization of those accounts. A memorialized account remains online, but no one can log in to it as that user, which prevents the account itself from posting spam messages.

On Facebook, there’s a middle ground for account access: users can appoint a “legacy contact” who has the ability to manage your profile once memorialized, kind of like a group administrator — but they can’t post as “you.” The “legacy contact” option is available in the security settings menu on Facebook.

Facebook and Instagram also try to prevent memorialized accounts from algorithmically resurfacing in a way that might be disturbing to those who knew their late owners. It’s worth noting, though, that memorializing an account doesn’t fix every digital afterlife problem.

Twitter provides its users with fewer options for deceased account holders. Immediate family members, with verified proof of death, can request an account deactivation. In limited circumstances, Twitter says it will “remove imagery of deceased individuals” upon request. For those family members who would prefer to have the content on a Twitter account remain available after death, the only option is really to leave it as-is, just like any other account.

Of course, some make arrangements before death to hand over access to their accounts to a trusted friend or family member – and there’s plenty of advice on the Internet for how to do this.

An account handover is what happened with Roger Ebert’s popular Twitter account, which is still active after his death in 2013. Although most of the tweets pertain to rogerebert.com – a site that is also still active after Ebert’s death – Ebert’s wife Chaz accidentally sent this tweet, meant to be a DM, in 2015:

“This account is eating up all of my data and I want you to cancel it. Please respond”

Chaz, through the Ebert account, later indicated that her late husband had laid out some specific instructions for the use of his Twitter account after death.

The jarring Ebert tweet was simply a mistake. So what happens when a Twitter account like Carr’s is posting spam? A spokesperson for Twitter declined to elaborate on the specific circumstances under which Twitter might intervene to stop spam an a deceased user’s account, and directed us to their general information on restoring a hacked account.

The spokesperson also declined to comment on whether there are any special considerations for verified accounts connected to the identity of someone who has died, like Ebert’s and Carr’s, or whether Twitter was considering developing its own version of the memorialization option available on Facebook and Instagram.

In Carr’s case, plenty of journalists – i.e. the sort of people who would know exactly how to contact Twitter efficiently — brought the spam to the site’s attention, and the account was restored pretty quickly.

The spambot wasn’t the first ghost. Before the Carr account began to tweet, his old email address was apparently sending spam to his contacts. “There is almost nothing worse than getting spam from David Carr’s old/hacked email address,” Clara Jeffery, the editor-in-chief of Mother Jones tweeted. Carr’s Twitter account was restored within hours.