Feel free to skip this part and go to below the line separator:
I hate that I can’t use em dashes and hyphens anymore. Or that I have to either use two or four examples as AI often outputs three.
And even then, I still can’t tell if something has been written by an AI or not as they could’ve instructed the output to;
- identify and remove those artefact’s and new ones by researching the latest identifiers we post about online, themselves or asking the model to do so as part of the task
- to write in the tone of an author or journalist or even myself based on a WhatsApp or diary export, an amalgamation of all
It’s easy to spot the obvious, it’s much harder when the trails start getting covered and doubt begins to creep in.
———
Back to the meat of it:
I’m at the point now where the content doesn’t matter but the intention. Like political speak, what you say and what you mean can be vary by orders of magnitude.
One day soon I’m sure every user will have their own personalised UI built from preferences and data collected about them to tailor those interfaces and content to how they digest it best.
Is it possible that then governments, service providers and digital products become nothing more than API’s we allow agents to pull data from and allow them to interpret that data how they like within the constraints of “you must say this for legal purposes” or “must include X”?
As an example, think skeuomorphic design. A design method meant to help you understand to a digital function via something you already understand. The trash bin. An analogy.
Or social media marketing. It’s targeted to a demographic and tailored to you because of the data they’ve collected. We are not far from the demographic being a person not a range.
———
My core point/question is: are we heading towards Personalised Analogical User Experiences?
[link] [comments]