Desire2Learn has announced its latest grab: Knowillage Systems, Inc. Knowillage makes LeaP, an artificial intelligence that can sit on top of a learning management system. LeaP tracks individual K-12 grade student’s performance through online assignments and quizzes, and grows smarter with everything it sees. It then tailors the student’s lessons to them, offering different types of readings or practice problems if the student is struggling.

Basically, it’s a robot teacher, albeit one that operates through a computer. ...

It’s a little sad that our educational system is so strapped that we need machines to give students individualized care.

But wishing we had more teachers doesn’t make it so... Carmel Deamicis, "When Your Teacher is a Robot", Pandodaily, September 9, 2013.  

Cappo, Marge, Fred D'Ignazion et al. Technology, trends, and gizmos: a timeline for the '90s... and beyond. Technology & Learning. 11.1 (Sept. 1990): p92.*

When I read about how we use educational data in schools, some questions come to mind: 

1. Is the technology a decision-making tool or a diagnostic tool? Who is the better diagnostician: the doctor or the database? Who can create a better IEP: a teacher or a database? Or is the best diagnostician the doctor or the teacher who skillfully uses a database?
2. Why haven't we been able to apply the concept of mass customization to education? Traditionally, one had the choice of buying blue jeans either off the rack (mass produced and cheap) or tailor-made by hand (customized and expensive). But as I remember, one savvy manufacturer combined the two by having individuals send in their specific measurements which were then input into a program that directed a robot to make a unique pair of jeans with exactly those measurements. As the opening quote suggests, artificial intelligence programs that work with data mining and CMS systems might make this a reality. (The love child of Moodle and Viewpoint in our district.) 
3. If want to use data to help individualize students, what is the most valuable information to record, organize and analyze? I see data-mining being done primarily to evaluate schools, teachers, and groups of students - not individual students. Given that our data repository primarily holds state testing and NWEA MAPS data from tests given one or twice a year, there is insufficient granularity to use the analytics to design programs for individual students. (We can say that a building's 4th graders are not doing well in math, but does that help the specific individual who is struggling?) Do we need to stop investing in and fussing about state/national norm-reference test scores, and instead figure out how to record performance data on individual skills as measured by formative assessment tied to local learning outcomes?** Seems more useful for day-to-day guidance.

I do think technology will play a role in helping individualize instruction. But at the same time, I do not want to lose the human element in education. I want a caring teacher, wisely using data, to make the final call on how my grandchildren are performing, what they need to work on, and the resources and activities needed to get them there. I want their educations to not just be individualized, but personalized, and I don't think a robot teacher will ever be able to do that.

Image source

*This is an article well-worth finding. (I had a copy inter-library loaned to me since I could not find the graphic version online - the graphics contain the bulk of the information, and they're lost in the text version available from full-text databases.

** I took an educational jargon-enhancement pill this morning.