Facebook CEO Mark Zuckerberg has finally built his own artificial intelligence assistant this year, named after the AI in "Iron Man," Jarvis. Zuckerberg's new housemate can control different appliances, play music, recognize faces, and most remarkably --- entertain his daughter, Max.

Zuckerberg spent 100 hours creating Jarvis, the virtual assistant. It understands spoken commands and text messages, according to his 3,000 word Facebook post last Monday.

Jarvis' skills include operating the toaster, adjusting the home thermostat, and turning on the lights. In addition, Jarvis can text Zuckerberg his visitors' images who either stop by during the day or open the front door if that person is recognized. It can also tell whether Zuckerberg's 1-year-old daughter, Max, has woken up --- so it can either start playing music or a Mandarin lesson, as what Zuckerberg mentioned in his Facebook post.

In his tongue-in-cheek video posted last Tuesday on Facebook, Zuckerberg has shown some of Jarvis' skills at work --- some of which were Jarvis (voiced by Morgan Freeman) telling him about his daughter waking up, greeting his daughter, and if she's about to practice Mandarin.

According to Zuckerberg's post, the year-long project was to learn about the state of artificial intelligence and an opportunity to experiment with cutting-edge technology at a period when voice-activated assistants such as Google Home and Amazon's Echo are having global popularity.

Zuckerberg mostly ask its virtual assistant, Jarvis, to play music --- making Jarvis look at his past listening patterns, so it can learn which music to play or not. Not only that; he also frequently asks more open-ended requests rather than specific questions, so Jarvis can learn about classifications and recommendations based on queries. He sees this as a big opportunity since no other commercial products as of today can do the same, according to The Washington Post.

Zuckerberg then plans to create an Android app for Jarvis and connect it to more appliances around the house. According to him, the ultimate challenge was to build a system that can learn completely new skills by itself.