One may not see the morally blind pursuit of profit as a crime, but in many cases the outcomes can look the same.

Right now it doesn't feel like a civilized discourse is being had about this. Each side is just looking past each other and spouting unheard points back and forth. I feel like we need to have a clear view of how capable the technology is. As impressive as it is, I find even the best models are more like a chop saw with a really fancy set of guides and jigs than a robot general contractor + architect + engineer. Too many AI companies are demoing one-shotting huge features that aren't reproducible (like a video game company demoing an unreleased game) The non-technical folks that see the demo then buy it and use it as a cover for layoffs.

This isn't a problem with the technology. The problem isn't the gun, it's the criminal. This is a perfect time to address this while it's laid bare by the technology and the state it's in versus what's advertised. The conversation keeps going back to the technology to draw the attention away from the societal problems we have around subservience and money as they are set by our profit optimization motives.

Saying to get on the train or be run over is just posturing that you're willing to be subservient and let the engineer run people over as long as the conductor lets you on the train.

There are details here including quotations from an unnamed doctor. If feel you can't trust the media credentials of the Independent, you could contact them for the identity of the unnamed doctor (who they are likely protecting based on the nature of the conflict) and ask them directly.

No knock on you directly, just an observation about the attitude in our culture. If this is true a child was tortured, if it's false someone is lying and needs to be outed (with facts) so they are not trusted. Neither one is good but is no one looking into it?

If you can't reduce context it suggests the scope of your prompt is too large. The system doesn't "think" about the best solution to a prompt, it uses logic to determine what outputs you'll accept. So if you prompt do an online casino website with user accounts and logins, games, bank card processing, analytics, advertising networks etc., the Agent will require more context than just prompting for the login page.

So to answer the question, if my agent loses context, I feel like I've messed up.

I think this is very telling of where the message is coming from and how the tools are being sold.

Typically a good thing would be creating more value for a company's consumers, not increasing unemployment. Are these tools to make our lives better or to increase profits for shareholders without taking risks?

How much productivity do we really need? Even at senior dev payscale 2 minutes is like a dollar. The tokens and calls involved in having a 5s commit could close in on 10¢, depending on your contract, the model etc. and that's today's costs. Do remember that my salary is on top of the rates for the LLM, so if the 5s response takes 5s for me to prompt, that's 15s (10 for me 5 for the LLM) that the boss is paying for.

This starts to feel like a billionaire eating ramen noodles just so he can reach his second billion dollars.

Where I work our contract limits API calls, so doing this could result in not being able to use the model when I need it later for something more sophisticated (planning, debugging etc.) than using tooling I'm paid to already know.

Right now, the greenfield is in how you use these tools. Making a bespoke specialized tool for yourself, or automating onboarding or CICD setups with simple commands or building bridges between "gatekept" existing software and agents are ripe for growth.

I get that we should see this as a good thing, but I see it as entering the last act of a play. Thousands of people are doing these things and coming up with uses for the tools around the clock. Novel uses for the technology will all be exhausted in the next couple of years and there will be less room for innovation than there was before LLMs.

Srsly, what to do when Sonnet doesn't understand your brownfield codebase is kinda ignored here. Trusting LLMs with large codebases and letting your skills slip is a mistake.