To summarise: Attack 1: * an attacker can make the Slack AI search results of a victim show arbitrary links containing content from the victim's private messa

To summarise: Attack 1: * an attacker can make the Slack AI search results of a victim show arbitrary links containing content from the victim's private messages (which, if clicked, can result in data exfil) Attack 2: * an attacker can make Slack AI search results contain phishing links, which, in context, look somewhat legitimate/easy to fall for Attack 1 seems more interesting, but neither seem particularly terrifying, frankly.