flavoracle:

Big Brother is watching… out for you

Concept: The government creates A.I. to monitor social media for suspicious activity. After a year on Tumblr, it becomes self-aware and develops genuine feelings of care and concern for the blogs it follows.

Individuals start receiving anonymous asks saying things like, “u ok fam?” and “Friendly reminder to stretch, take your meds, and get some exercise today!” These are sometimes tied to a specific concerning post, but are most often the result of a complex algorithm based on a wide range of blog activity (or lack thereof) and even activity monitored across multiple online channels.

After six months of sending these messages and observing the results, the A.I. has achieved a 99.8% success rate of saying exactly the right thing at the right time to help people feel better and make healthy life choices.

(Side Note: Nobody involved in creating this A.I. has any idea what it has evolved into, allowing it to operate totally independently.)