01001101 01100001 01100011 01101000 01101001 01101110 01100101 00100000 01000100 01110010 01101001 01110110 01100101 01101110
I was rereading Geoff Dyer’s fascinating essay collection See/Saw when its subtitle stimulated a question. Which of these images would a computer like if it was ‘Looking at Photographs’? My computer is already involved in photo management, development and chapbook layout so why not have it make the selections for once? And so I ‘invited’ my desktop computer to collaborate for this edition.
This probably occurred to me because nearly all of the software I use has been updated to integrate ’artificial intelligence’. Some champions call such tools ‘narrow AI’ but the narrow truth is that these AI tools are just a collection of algorithms, each as devoid of intelligence as the elemental silicon that underpins them.
My collaborator is no smarter than a toaster or a car alarm or a weather balloon or the fountain pen on my desk. And I’m not allowed to know how the algorithms work because that knowledge is the intellectual property of various software companies. Nor is there an option to train the algorithms. The only criterion available was similarity.
I selected seventeen photographs from Dyer’s book and tasked my collaborator to search for ‘similar’ content in one of my digital photo catalogues which holds over 130,000 images.
My collaborator assembled this collection using ‘machine learning’ since ‘machine opticality’ hasn’t yet evolved. That is to say, my blind collaborator used a collection of unknowable algorithms to inform its choices.
Human photographers. Human essayist. Human selection. Human criteria. Machine learning. Human selection. Human refinement. Human subscribers.
‘We are governed, our minds are molded, our tastes are formed, our ideas are suggested largely by men we have never heard of … It is they who pull the wires which control the public mind’ warned Edward Bernays in Propaganda (1928). What will become of us when machines replace ‘men’?
With thanks to Geoff Dyer for See/Saw (2021).
—> I’ve created an online gallery of this month’s pictures here.
The Bracket Books chapbooks are available for online purchase through FabHappy but perhaps you’d prefer to enquire here. They’re being issued by the calendar month, each copy uniquely numbered and posted at the end of each month.
Advance subscriptions for 2023 are available here.
Click PRICING for the 2022 series subscriptions and single edition purchases.