gb2d_hn

Had a play with the native app - it's a complete new set of keyboard shortcuts from the TUI. Don't really see the advantage.

Yeah that copy is pretty bad. My understanding (someone correct me if I'm wrong), but the herdr native app is an improvement over the existing TUI multiplexer, with a new API for facilitating comms between agents. Herdr is really useful if you've not tried it. I'll be giving this a go (I hadn't really been using agent comms between panes anyway, more just for workspace navigation). Seems you can think of herdr as a 'harness for your harnesses'!

Why would I buy a potato when I could grow one myself. Why do I give a busker money when I could play the real track on spotify. Why do we give anyone any money? Why don't we just do everything ourselves?

Your friends quote is so true. There's too many thin stories with empty padding around them to make an idea a novel!

Currently, for me the challenge has gone from can I build it, to can I build this and everything else I need, and structure it to be maintainable for the long term, so I can keep stacking bricks. This has found me leaning harder into IaC, use of and contribution to open source (for secure ownership, longevity of system and cost reduction) and structuring the development tooling around agents. That will be an ongoing process, but I feel that I'm at the point where I can move onto building what matters. I've already built a few of the tools I've dreamed of having for years (there's a couple of rough edges but it beats the tools I had before), and I think, like the author this will free me up to stay on the next step (building what matters) for good.

There's a couple in that collection - 'Understand' is about a brain injury patient that gets an experimental treatment and becomes super intelligent - what he does with that intelligence and humanities attempts to restrain him, and 'Division by Zero' is a mathematicians struggle to continue with life when a discovery undermines all of mathematics. It's a great collection.

I'm ashamed to say I've started both of these and not managed to finish :) I will try again (kiddo's tire me out and means Ted Chaings short stories are perfect!)

As I commented in another thread: I just finished a very short story by Ted Chiang - 'The evolution of Human science' it starts at the point where it's 25 years since humans contributed a scientific paper because 'meta humans' now handle all research, and communicate digitally such that humans can only get a low fidelity report of what the agents actually know, and the endeavour to understand has become like a best effort archeological like hobby for interested humans. The story assumes that the 'meta humans' will always act in humans best interests. I believe it's from 2002, it's incredibly precient (The stories of your life and others is the collection the story is in. Many of them are very relevant to modern AI actually).

You can solve problems empirically, but you can never guarantee you've hit all the edge cases the original code did. That's fine for non-mission critical systems, but would you be ok with your passenger jet onboard systems having been rebuilt from empirical observations?

I think the difference is whether agent availability in the future is guaranteed. I agree that no one had the whole picture before, but at least theoretically you could replace x employees with another workforce who in time could be expected to match the understanding and output of the original workforce. With agents, if capability is consolidated with a single organisation and that organisation changes it's terms of business, you cannot replace the army of AGI-agents with a human workforce. It's a single and total point of failure.

Thank you, I'll try and find those.

I think this could be summarised as alignment being conflated with intelligence - i.e. what seems unintelligent to some is just misalignment with their own values.

This just can't be sustainable. Ever increasing agent activity can only move to the point where it's literally impossible for humans to have a collective awareness of how the systems that support our lives actually work. How does anyone think this is going to end well? If as humans we loose control of vital systems, when they stop working one day (and they so easily could unless the agents are perfect and 100pct aligned with the general good of humanity (how do we even define that)) we could find ourselves without the ability to run healthcare systems and power grids, that have evolved so much such complexity that no human can intervene to fix them. I do believe that you can create feed back loops where probabilistic agents can be used to create deterministic checks and balances, and solve real problems for good. But it's a wild west at the moment and that is not what's happening in so many companies, and the solutions are not yet on the horizon. Even if the 'agents can handle it' ... where does that leave us? Unable to run our societies without puting everything through frontier LLMs? I just finished a very short story by Ted Chiang - 'The evolution of Human science' it starts at the point where it's 25 years since humans contributed a scientific paper because 'meta humans' now handle all research, and communicate digitally such that humans can only get a low fidelity report of what the agents actually know, and the endeavour to understand has become like a best effort archeological like hobby for interested humans. The story assumes that the 'meta humans' will always act in humans best interests. I believe it's from 2002, it's incredibly precient (The stories of your life and others is the collection the story is in. Many of them are very relevant to modern AI actually).