Beginnings

Back in Jan 2023 I somehow got it into my head that I should investigate or at least have a look at what all the fuss was about in relation to AI and Generative Image making. I didn’t understand it in the slightest and was less than impressed with the idea of instructing some new fangled software to make me an image of ‘X’. What was the point, I can draw, paint, print, photograph and work across a range of CGI software and digital compositing software. Why should i bother with AI? Somehow the wheel turned and I found myself thinking that I could tackle the beast head on and do something different with it. Being then rebel that I am it came to me that i should fashion prompts along the lines of creating problems for the AI software to solve. Otherwise it’s just same ol’ same ol’.

So down one rabbit hole after another to try and get to the bottom of how to go about setting my mischief into place. At the time the hot topic was intellectual property theft, mainly from photographers and digital artists. I knew enough to know that this was a bit of a dark horse and a furphy to boot, but it gave me the leg up that I needed. What kind of images / image data won’t have been picked by the web crawlers trawling online storage and social media platforms; in brief, what sort of images don’t exist. That got my imagination working and the first prompts were quickly being tested. The thing is, I want to be surprised, I don’t want the result to be expected or predictable and this is where I also found a working companion in GANN’s (Generative Adversarial Neural Networks). Despite the often (but not always) superior generative quality of software like Stable Diffusion, Dall-e and Mid Journey I found the results to be too predictable and often a bit lame when a particular ‘style’ was applied.

My journey begins with Wombo AI out of Canada. Here are a random selection of some of the earliest images from back in Jan 2023.

Just a little chaos

This is an unusual thing for me to do but under the circumstances, probably a necessity. The reason being that;

  1. EdConnect has no category for this type of complaint and
  2. There is too much information to permit submission via an online request. So reducing it to a URL makes that task a lot easier.

On the 9th of April ’23 I received this email from IDM_NoReply@det.nsw.edu.au informing me of the following. (see image) Under normal circumstances, and; if the information contained in the email was correct, there would be no problem. However, this mail notification is riddled with some serious errors that spell out loudly and clearly that certain processes within the DoE are basically up the creek.

I have been with the Dept of Education for over 27 years, (Visual Art, Photography & Visual Design). Of those, between 1997 and 2020 were as a full time member of staff. I retired from full time teaching in May 2020 after undergoing a quadruple bypass in December 2018. I retained my employment status (unbroken) with the Dept and have worked as a relief teacher on and off for the last 4 years. My DoE email address has been the same for the last 27 years.

Unravelling the mess.

“During your time as a casual staff member of the NSW Department of Education, an email account was provided to you for carrying out business related duties on the basis that it would expire fifteen months from the date of your last pay”

So firstly; no conditional email account was provided to me. I simply continued using the one I had been using for the last 23 years. In the email from the DoE (May 2020) notifying me of my “approval to teach’ as a relief / casual teacher no mention was made that my current email address would at any time expire under the conditions mentioned in the email in question. Had I been resigning from the DoE it would be a foregone conclusion that my DoE email would cease to be operational along with access to all other Departmental portals. However I have not resigned nor requested a separation from the DoE.

“Accordingly, you are notified that your email account will expire and be removed on 9 May 2024. Email accounts that remain active after a staff member has left an organisation create security risk that the Department is required to address under its Information Security Management System (ISMS) and obligations to the NSW Digital Information Security Policy (DISP).”

Now, 15 months from the 9th May 2024 is 9th Feb 2023. The problem here is that I was working at Nepean Creative and Performing Arts High School on a Permanent Part Time contract which I did not relinquish until 18th July 2023 due to complications from ‘SARS COV-2’ (infections x 2) and ‘Influenza A’ courtesy of my work environment.

” Email accounts that remain active after a staff member has left an organisation…….” What on earth are these people doing or not doing? as the case may be. All the correct information is available under ESS in the SAP Portal. The information there is populated by operatives within the DoE so how on God’s earth do mistakes such as are outlined in this post, come into being.

I offer two suggestions;

  1. People don’t know how to do their job
  2. AI assisted software is being used, in which case, heaven help us.

Given that I have not;

  1. Resigned from the DoE
  2. Applied for seperation

There is no justifiable reason to remove my current email and the contents thereof. To do so would also deny me access to all employee services to which I am still entitled until i do at some future date terminate my employment with the Department of Education.

For those in the Dept to whom this post is addressed, FYI, my last payslip is dated 04/01/24. For which I have screenshot and downloaded PDF. Which puts the conditional termination of my DoE email upon resigning (if no further relief work is undertaken) at 04/04/25.

Highlights of my working history with the Dept of Education can be viewed here;

Lost Horizons

Addendum

Issue resolved (megasaga) 26/4/24

Artificially Intelligent? or just a ‘clone ranger’

It’s been a while, a whole year nearly since posting to this site. A lot has happened some of which is worth talking about. It’s been the year of generative image making and ChatGPT. Myths and legends abound and for me it was a year to do some myth busting.

ChatGPT was my entry point. Through some extensive dialog with educators who had concerns about student usage and the supposed advantages thereof I found myself doing some system testing to see the extent of the application of generative text and to better understand the process. Leaving aside the standard “write me an essay about…….” I went straight for the throat and entered questions directly from HSC exam papers (relevant only to teachers in NSW Australia) What i discovered was that that ChatGPT struggled with contextualising the nature of the question and the response. Firstly given that most exam or essay questions are not questions but instructions; questions generally start with what, when, where, why, who, how etc and not verbs such as discuss, investigate, analyse, compare, describe, assess, clarify, evaluate, examine, identify, outline etc the responses were consistently very average; if you were to scale them most would fall into the ‘C’ range. So; no advantage to be had here. This then led to ‘why does this happen’ and to understand that I realised i had to look at ‘how’ this all happens. That led to looking at decoding and encoding text.

Machines, so it appears do not ‘read’ text as we do. Whilst the eyes and brain are involved in a decoding and encoding of the letter shapes and their relationships, machine learning involves encoding text something like this;

‘\x74\x27\x73\x20\x62\x65\x65\x6e\x20\x61\x20\x77\x68\x69\x6c\x65\x2c\x20\x61\x20\x77\x68\x6f\x6c\x65\x20\x79\x65\x61\x72\x20\x6e\x65\x61\x72\x6c\x79\x20\x73\x69\x6e\x63\x65\x20\x70\x6f\x73\x74\x69\x6e\x67\x20\x74\x6f\x20\x74\x68\x69\x73\x20\x73\x69\x74\x65\x2e\x20\x41\x20\x6c\x6f\x74\x20\x68\x61\x73\x20\x68\x61\x70\x70\x65\x6e\x65\x64\x20\x73\x6f\x6d\x65\x20\x6f\x66\x20\x77\x68\x69\x63\x68\x20\x69\x73\x20\x77\x6f\x72\x74\x68\x20\x74\x61\x6c\x6b\x69\x6e\x67\x20\x61\x62\x6f\x75\x74\x2e\x20\x49\x74\x27\x73\x20\x62\x65\x65\x6e\x20\x74\x68\x65\x20\x79\x65\x61\x72\x20\x6f\x66\x20\x67\x65\x6e\x65\x72\x61\x74\x69\x76\x65\x20\x69\x6d\x61\x67\x65\x20\x6d\x61\x6b\x69\x6e\x67\x20\x61\x6e\x64\x20\x43\x68\x61\x74\x47\x50\x54\x2e\x20\x4d\x79\x74\x68\x73\x20\x61\x6e\x64\x20\x6c\x65\x67\x65\x6e\x64\x73\x20\x61\x62\x6f\x75\x6e\x64\x20\x61\x6e\x64\x20\x66\x6f\x72\x20\x6d\x65\x20\x69\x74\x20\x77\x61\x73\x20\x61\x20\x79\x65\x61\x72\x20\x74\x6f\x20\x64\x6f\x20\x73\x6f\x6d\x65\x20\x6d\x79\x74\x68\x20\x62\x75\x73\x74\x69\x6e\x67\x2e\x20’

The above is the first paragraph of this post encoded in UTF-8 Hex code

Encoded in UTF-32 it looks like this; u+00000074u+00000027u+00000073u+00000020u+00000062u+00000065u+00000065u+0000006eu+00000020u+00000061u+00000020u+00000077u+00000068u+00000069u+0000006cu+00000065u+0000002cu+00000020u+00000061u+00000020u+00000077u+00000068u+0000006fu+0000006cu+00000065u+00000020u+00000079u+00000065u+00000061u+00000072u+00000020u+0000006eu+00000065u+00000061u+00000072u+0000006cu+00000079u+00000020u+00000073u+00000069u+0000006eu+00000063u+00000065u+00000020u+00000070u+0000006fu+00000073u+00000074u+00000069u+0000006eu+00000067u+00000020u+00000074u+0000006fu+00000020u+00000074u+00000068u+00000069u+00000073u+00000020u+00000073u+00000069u+00000074u+00000065u+0000002eu+00000020u+00000041u+00000020u+0000006cu+0000006fu+00000074u+00000020u+00000068u+00000061u+00000073u+00000020u+00000068u+00000061u+00000070u+00000070u+00000065u+0000006eu+00000065u+00000064u+00000020u+00000073u+0000006fu+0000006du+00000065u+00000020u+0000006fu+00000066u+00000020u+00000077u+00000068u+00000069u+00000063u+00000068u+00000020u+00000069u+00000073u+00000020u+00000077u+0000006fu+00000072u+00000074u+00000068u+00000020u+00000074u+00000061u+0000006cu+0000006bu+00000069u+0000006eu+00000067u+00000020u+00000061u+00000062u+0000006fu+00000075u+00000074u+0000002eu+00000020u+00000049u+00000074u+00000027u+00000073u+00000020u+00000062u+00000065u+00000065u+00…………etc, etc

So when a prompt is entered into ChatGPT it is first encoded so that it can be read. The response from ChatGPT is likewise scripted in machine language and then decoded into text. How does it work?

It’s primarily a predictive model, that predicts sequences based on learnings from ‘other encodings’, because that’s what the neural network reads. When this is understood a lot of the ‘myth-understandings’ about machine learning are to some degree dissolved. The better the quality of the language structure of the texts that neural networks are trained on the better the likelihood of a cohesive albeit somewhat standardised, (bearing in mind the encoding and decoding sequence) response.

to be continued @ STAGESIX

  • Code points are numbers that represent Unicode characters. “A code point is the atomic unit of information. Text is a sequence of code points. Each code point is a number which is given meaning by the Unicode standard.”
  • Code units are numbers that encode code points to store or transmit Unicode text. One or more code units encode a single code point. Each code unit has the same size, which depends on the encoding format that is used. The most popular format, UTF-8, has 8-bit code units. @https://www.coderstool.com/unicode-text-converter
  • Code points are converted into ‘tokens’. The relationship between tokens is calculated in relation to the ‘prior learning’ in the LLM.

Laneway

IMG_6357

Hipstamatic 303 @ Laneway Sandwich and Espresso, Leura.

Views from windows; typically unforgiving in that what you have is for the most part what you shoot. Finding the sweet spot in what’s in front of you calls for continued culling of non essentials. The ‘day to day’ fascinates me, finding the edge in the ordinary moment. However…….

……….Great coffee, food and service 🙂

IPPAWARDS 2015

A big congratulations to Michal Koralewski, David Craik and Yvonne Lu for picking up the top spots in this years iPhone Photography Awards ( IPPAWARDS ). Fellow Australian and Sydney-sider Nicky Ryan picked up third place in the ‘Seasons’ category along with a number of honorable mentions. I was fortunate enough to be selected for 3 Honorable Mentions in ‘Still Life‘, ‘Portrait‘ and ‘Others‘. see below

The Impulse to Draw @ Adobe Slate

Having been involved in the beta of Adobe Luca, it was a no brainer that when Adobe released ‘Slate’ on the App Store that I should dive into this with renewed enthusiasm. ‘Slate’ is a significant improvement on the beta and in hindsight I found that my earlier publications needed revisiting and that’s still ahead.

The latest iteration of SLATE brings ‘Glideshow’, a new feature that facilitates image dissolves with text gliding over the top. Kinda nice. Haven’t used it yet but here is something of a recent offering. Life at art school and beyond.

IMPULSE TO DRAW