       In the Stanford A.I. report, the researchers proposed many promising fields for A.I.s application in near future, including transportation, healthcare, public safety etc. Among those fields is education, which has seen many budding innovations driven by the increasingly powerful A.I. technologies in its field. Backed by examples such as emerging intelligent tutors, explosion of Massive Open Online Coursers and learning analytics, the report reached a cautious and optimistic conclusion in Broader societal consequences section that more online education resources accompanied by A.I. technology, may help further democratizing education and benefit student from countries where education is difficult to obtain, if they are equipped with tools to access them.
       Although it should be acknowledged that the recent burst of online education resources has acted as a positive force toward democratizing education, a more holistic observation may raise some doubts about their effectiveness in practice. The optimistic viewpoint on online education resources augmented by A.I. technologies such as MOOCs and language learning apps is largely based on the assumption that digitalized educational material can be reproduced and disseminated more easily while A.I. technology like NLP make the automation of such as responding and grading possible and together they makes teaching and learning a more scalable experience and therefore more accessible for students who previously lack education opportunity.
       However, there are still other steps to be taken for the students in need to really benefit from online resources after course materials and A.I. tutors are put online. First, student should have the ability to really access and utilize these resources. Here lies the issue of digital literacy and Internet infrastructure development. If might be safe to say that the students who lack access to traditional education and therefore are most in need of online education are those who have disadvantaged socioeconomic background and more likely to live in developing countries and less urbanized area. This means they may lack the necessary tools to access these resources like computers and reliable Internet connection. These students may also be less familiar with the use of online resources and hard to utilize these resources in practice, given these online learning resources usually put an emphasis on self-learning. This self-regulation feature of online education resources also leads to the problem of students motivation. Students who lack formal school education may have less chance to form good study habits and such as setting their own goals and time management, which are more important in online learning. Therefore, students who are most in need of more accessible education are put to an disadvantaged points in online education. Some research on Massive Online Open courses confirmed these concerns. A survey conducted by G Christensen et al (2013) shows most MOOC sample students in have already had post-secondary degree. The education level of MOOC students far exceeds that of the population in the corresponding area, which is especially significant in developing countries and more than 40% of surveyed students report they take the course to advance their career. These results indicate instead of a tool to help less educated individuals, online learning is more likely to serve people who already have high education attainments. To solve these barriers inevitably raised due to the very nature of digital and online education requires approaches from socioeconomic aspects such as building infrastructure. The effectiveness of the educational A.I. technologies discussed within the scope alone may not achieve as significant a goal as the researches expected, especially in the perspective of improving education equality.
	 
