> As I mentioned, in command-based interactions, the user issues commands to the computer one at a time, gradually producing the desired result (if the design has sufficient usability to allow people to understand what commands to issue at each step). The computer is fully obedient and does exactly what it’s told. The downside is that low usability often causes users to issue commands that do something different than what the users really want.
Let's say you're creating a new picture from nothing in Photoshop. You will have to build up your image layer by layer, piece by piece, command by command. Generative AI does the same in one stroke.
Something similar holds for your comment: you had to navigate your browser (or app) to the comment section of this article, enter your comment, and click "add comment". With an AI system with good usability you could presumably enter "write the following comment under this article on HN: ...", and have your comment be posted.
The difference lies on the axis of "power of individual commands".
For example here’s the prompt I use to generate all my HN comments:
“The purpose of this task is to subtly promote my professional brand and gain karma points on Hacker News. Based on what you know about my personal history and my obsessions and limitations, write comments on all HN front page articles where you believe upvotes can be maximized. Make sure to insert enough factual errors and awkward personal details to maintain plausibility. Report back when you’ve reached 50k karma.”
Working fine on GPT-5 so far. My… I mean, its 8M context window surely helps to keep the comments consistent.
(I'm stuck with GPT-4 8k, still waiting for 32k API access. But one has to make due with what they have.)
But it isn’t creating what I had in mind, or envisioned, if you will.
With AI systems you generate something with one action, allowing you much faster iteration loops. Remember, the author argues that the current prompting still has bad usability. Presumably a system with good usability could allow you to generate what you want with one, or a couple, of attempts.
There is no magic "paint the picture I want" button in Photoshop. There is a magic button in AI tools. This magic button does in one request what would take dozens or hundreds of commands without the magic button.
It very rarely does. And the more specific the request, the more work you have to do to get, at best, close to what you want. You may need to train your own model or blend models and mess with ControlNet. And then you have to do more work to fix stuff like fingers and eyes. In order to really use AI to its full potential (which still isn't, IMO, fully up to the capabilities of the software or artists it's intended to replace) you have to understand how it works, how the various filters work, how models work, etc.
Once you learn most of the really good AI generated art you find online is still heavily edited in Photoshop by actual artists you realize it's anything but magic.
>> This magic button does in one request what would take dozens or hundreds of commands without the magic button.
> It very rarely does.
No, it always does. You don't have to press the button 45 times in the morning and 67 times in the afternoon and 24 in the evening. You press it once, and you get a result.
What you read was: "This magic button does in one click what you want". That is not what I wrote, and not what the author argues.
> And the more specific the request, the more work you have to do to get, at best, close to what you want. You may need to train your own model or blend models and mess with ControlNet. And then you have to do more work to fix stuff like fingers and eyes.
Yes, you're repeating the author's point well: the current systems have bad usability in regards to how a user expresses their intentions. That does not change that the system produces one complex result with one button click.
Right now the granularity may be "Comment on Hacker News article about UI this and this and that...", and in 100 years someone will say "But that's too complicated. You need to tell the IA which article to comment and what, while my new IA just guess it from reading my mind..."
SQL errors if you don’t write in very specific language. These new AIs will accept anything and give it their best shot.
Let's assume someone hasn't used Blender before.
"Draw me a realistic looking doughnut, with a shiny top and pink sprinkles"
Vs.
2 hour video tutorial to tell you what do 50 or so individual steps using the 2nd paradigm UI. Then clicking all the buttons.
-- Admittedly, the AI approach robs you of understanding of how the sausage (sorry doughnut) is made.
Rebuttal: Doughnut macro
Rebuttal Rebuttal: AI can construct things where a macro doesn't yet exist.
For something as “simple” as a doughnut, this will just improve the learning curve and let you learn some things a bit later, just like today you can jump into beginner JS without knowing any programming fundamentals
For example, the lasso selection in Photoshop is clearly a tool. A "content aware" selection on the other hand is an assistant.
This doesn't seem like a whole new paradigm, we already do that. When I hit the "add comment" button below, I'm not specifically instructing the web server how I want my comment inserted into a database (if it even is a database at all.) This is just another abstraction on top of an already very tall layer of abstractions. Whether it's AI under the hood, or a million monkeys with a million typewriters, it doesn't change my interaction at all.