Mental health professionals (MHPs) tell us that sometimes community members contact them in cases of social media postings that raise concern of suicide so that the MHP can reach out to the individual at risk. Unfortunately, it is often the case that such posts do not receive any reply. The experience of failing to get a response after disclosing thoughts of suicide could exacerbate feelings of isolation and worthlessness, potentially increasing risk of suicide. This situation motivates the need for a safety net which could detect cases of public social media posts that indicate crisis to ensure that they do not go ignored. Previous work by De Choudhury (2013), Jashinsky (2013), and others has shown that machine learning methods can detect signals from social media that indicate depression and risk of suicide. ARKHumanity is new technology which targets the language of depressed individuals who express their thoughts and disappointments through the social network, Twitter. It searches and filters public tweets to identify language that leads to self-harm actions using a machine learning algorithm. This technology also has a front end graphical user interface that helps clinical psychologists, behavioral health experts, and trained volunteers to connect to those who are in crisis by sending them an appropriate reply with relevant resources. Currently there is no technology like this. This interactive presentation will share this new innovative technology, share the preliminary results of the usability study and begin the discussion on how innovation can transform our society to save lives.