Safeguarding AI against ‘jailbreaks’ and other prompt attacks
Safeguarding AI against ‘jailbreaks’ and other prompt attacks by Vanessa Ho (() => { let $iframe, $closeBtn; const $player = document.getElementById(‘player-67bde5499a133’) ; const $playerPlayBtn = document.getElementById(‘player-btn-67bde5499a133’) ; const $playerPlayTitleBtn = document.getElementById(‘player-title-btn-67bde5499a133’) ; const $playerText = document.getElementById(‘player-text-67bde5499a133’) ; function onKeyDown(e) { if (document.activeElement !== $closeBtn) { $iframe.focus() ; } } function initIframe(e) { $iframe = document.