This is a most excellent place for technology news and articles.
We (my company) are trying to create agents that read a story and translate that into prompts, then execute said prompts, then review the output. The only piece missing is accepting the merge.
I'm not anti-AI, but a human needs to be involved at every step because a minor mistake made at the first step will amplify through the agentic pipeline.
A human should review every single thing that comes out of AI — especially if it is to be fed back into AI.
Some quotes that resonated with me:
In any previous era, the quality of a piece of work was a more or less reliable signal of the competence of the person who produced it. A novice essay read like a novice essay; novice code crashed in novice ways. AI has severed that relationship.
The skills of producing work and judging it were deliberately distinct, but accomplishing the work itself used to teach the judgment. The first skill now belongs, in large part, to the machines. The second still belongs to us, though fewer are bothering to acquire or utilize it.
The slowness was not a tax on the real work; the slowness was the real work. It was how the work got good, and how the people producing the work got good
The current generation of agentic systems is built around the premise that the human is the bottleneck — that the loop runs faster and cleaner without the awkward delay of someone reading what is about to happen and deciding whether it should. This is, in a great many cases, exactly backwards
