Comment by shiandow
13 hours ago
LLMs seem quite successful when considered something like a natural langiage interface, but expecting intelligence seems a step too far. For one they do not learn, at least not online, and that is a somewhat important requirement for truly intelligent behaviour.
Arguably programming is as much learning as it is writing code. This is part of the reason some people copy an entire API and don't realise they're not so much building useful code as building an understanding.
In some sense, programming is about figuring out which algorithms are a fitting metaphor for business problems. By programming, you are building a model of the business problem and a model of its solution. Most of the non-programmers who are in positions of authority (managers, CEOs, even some CTOs), do not understand that this is what programmers do. From their point of view, the authorities come up with a "strategy", after dozens of meetings, and give the programmers vague instructions based on the strategy, and programmers turn those instructions into code that does something somewhere, usually after finding ways to avoid bad or unfeasible ideas, while still complying with the instructions.
To them, an LLM is indistinguishable from a programmer. From the point of view of authority, progress happens one meeting at a time. The reality is that there is a pyramid of experts beneath the authorities, that keep everything running smoothly, in spite of the best attempts of the authorities to demolish the foundation of the pyramid by "helping".
EDIT: to end on a positive note, it does not have to be this way. We just have to be willing to understand _how_ the organization we are a part of actually functions. And that means actually being curious instead of merely authoritative. I understand that curiosity is hard to maintain when you swim with sharks, so maybe don't swim with sharks.
Funny thing is that from all places, banking and finance are some of the ones that more closely understand software not only as a tool for doing business, but as something that informs and define the business itself.
I once worked in a big bank where basically 70% of their C-suite and upper management were engineers (not only SE, but civil, electric, naval, etc) who had years of experience in IT. The rest were lawyers and a couple economists.
And I am not even talking about the worlds of quant and hft.