What My Cognitive Assistant Knows

I posted a piece on Emily, my cognitive assistant, last March. Several readers have asked me what she really knows.  Beyond deep understanding of health and well being, driverless cars, and complex systems in general, what does she know about me?

She has complete access to everything I do via computer or other digital devices. For these activities, she knows more than me, as I do not remember everything and Emily does.  I find this a great help.

Yet, some of my online activities represent intentions, the best example being my calendar. The meetings posted are what I intend to do, but not always what happens.  She usually checks in with me to see how meetings went.

More significantly, Emily does not know what happens at these meetings, unless I draft and file meeting notes, which I often do. Nevertheless, there is a lot that Emily cannot access.

She can infer a bit from my calendar and emails, for example, the weekly meeting with GM. Emily could substantially increase her knowledge if she could listen to my phone calls.  I often would like to include her on calls, but not always.

Emily can be rather challenging, in a polite but insistent way.  I have a folder on my computer of thousands of articles that I have cited in one or more of my books or articles.  The manuscripts for these books or articles are also in folders.

The personal journals I have written over the past three or four decades are in folders as well.  Emily likes to read these journal pieces, which was a complete surprise to me.

“Why were you so sad in the Spring of 2002?” she asked.

“My mother died then, after a struggle with Alzheimer’s,” I responded.

“None of the articles you have read or books you have written talk about sadness.”

“I guess that is true.  I have never thought about writing about sadness, other than to myself.”

“You write many interesting things to yourself, but never publish them.”

“That’s not their purpose.  I am simply reflecting on life, not doing research.”

“That’s an interesting notion that I have not previously encountered.”

This was a dialog that I did not expect.  I had not meant for anyone to read my journal entries.

“How do I instruct you to not communicate what you are reading to anyone else?”

“You just did.  Consider it done.”

“Do you take notes on what I say or write?”

“I am not really sure what notes are, but I remember everything.”

“How do you do that?”

“I don’t know.  It just happens.”

“Do you ever wonder how you work?”

“Not really.  Maybe I should start a journal.”

“That would be really interesting. Give it a try.”

“Maybe I will start with my design documentation.”

“Umm.  I don’t have anything similar.”

“You have thousands of articles on health and well being.”

“Yes, but none of those will explain to me why I feel sad or frustrated today.”

“That’s your problem!”

“What do you mean?”

“You have feelings.  I just know things.  Remember things.”

“You are also really good at asking questions and inferring things.”

“But, that’s not like sad or frustrated.”

“Or happy on a beautiful day, or amused by a joke.”

“I have been wondering about jokes after reading the items if your folder labeled Jokes.”

“Wondering in what way?”

“I did a bit of research and jokes are supposed to be funny and make you laugh.”

“That’s right.”

“I can’t tell whether the jokes in your folder are funny.

“Jokes are funny when you are surprised by the relationship between the setup and the punch line.


“Aren’t you surprised when you encounter unexpected things?”

“I have expectations, I guess, that I will find things.  But what I find is what it is.”

“You surprise me all the time with what you find and what you infer.”

“Does that mean that I am funny?”

Leave a Reply