I’ve tried to shedding much attention on OpenAI CEO drama. But after hearing whiffs of vapor about some Q star thing and the Amazon rolls out some other AI hooplah named… Q there is only one recourse.
That’s right, I haken back to my childhood adoration of the original James Bond flicks (I am solid Commery all the way down).
I remember having a plastic James Bond briefcase full of gadgets and tha that shot a rubber dart from the side. I still have a toy Aston Martin that at the press of a button ejects a plastic figure through the roof.
And all that cool technology device weaponry is due to the natural intelligence of Quartermaster aka Q.
I could not resist temptation over lunch to quickly rollout another iteration of a fun random generator housed under Edtechaphors named for the Edtech metaphor mixer inspired by Sir Metaphor Martin Weller. I fiddled with the code Martin originally found to create a Generator for Martin which I’ve made a handful of variants, the newest being Everyone is Rolling out an AI Named Q
Without too much technical spin, the generators are set up with word list arrays that can loosly be swapped to fill in blanks of sentences. You can see it if HTML and Javascript does not make you squeamish. One array is product and platform names (that’s where Anaconda Enterprise came from), and a few other arrays of various AI jargon like “neuromorphic computing”.
The jargon came from sources like Google’s Machine Language glossary, lists of platforms, and assorted lists of terms
What is fun about this generator code is that pulling from random word arrays is just one level, there are also an array of different sentence patterns each with a few blanks that can be filled with random words from the arrays.
The sentence patterns I used were lifted from a few movie quotes selections
“Need I remind you, 007, that you have a license to kill, not to break the traffic laws.”
various quote sites
“Explosive alarm clock, guaranteed never to wake up anybody who uses it.”
“I answer directly to m. I also have a mortgage. And two cats to feed.”
“Don’t touch that! That’s my lunch.”
“I can do more damage on my laptop sitting in my pajamas before my first cup of earl grey than you can do in a year in the field.”
“There are only about six people in the world who could set up fail-safes like this.”
“It’s called the future, so get used to it.”
My Q bot can spit out gems like:
There are only about six people in the world who could set up a generative adversarial network like this.
Don’t touch mini-descent! That’s my activation function.
I can do more damage on my positional encoding sitting in my pajamas before my first cup of backpropagation than NVIDIA can do with translational invariance in the field.
It’s called LaMDA, so get used to it.
Need I remind you, TensorFlow, that you have a license to chain-of-thought prompting, not to break the bounding box.
https://cogdog.github.io/edtechaphors/q/
What does this achieve? Not much beyond my own amusement, and also creating something from a bit of remixing of previous efforts and tweaking OPC (other people’s.code). And ChatGPT et al had no role at all.
As Q would say to that silly chatbox screen:
“You’re Not Here To Think. You’re Here To Do What I Tell You.”
The original Q
Featured Image: Zero Generative AI used Here. james-bond-q-gadgets flickr photo by tnssofres shared under a Creative Commons (BY) license modified with Photoshop Palette Knife filter overlain with single frame image from Amazon Q announcement screen video.
