6 ms·
This thing is doing planning and ascending the task management ladder. It's not just spitting out code anymore.
by cs391231 2y ago
This thing is doing planning and ascending the task management ladder. It's not just spitting out code anymore.
- fsloth 2y agoSure. But the added value of SWE is not ”spitting code”. Let’s see if I need to calibrate my optimism once I take the new model to a spin.
- parasubvert 2y agoAI Automated planning and action are an old (45+ year) field in AI with a rich history and a lot of successes. Another breakthrough in this area isn't going to eliminate engineering as a profession. The problem space is much bigger than what AI can tackle alone, it helps with emancipation for the humans that know how to include it in their workflows.
- hakanderyal 2y agoYes, and they will get better. Billions are being poured into them to improve. Yet I'm comparing these to the problems I solve every day and I don't see any plausible way they can replace me. But I'm using them for tasks that would have required me to hire a junior. Make that what you will.
- noshitsherlock 2y agoYes, if "efficiency" is your top concern, but I'd much prefer working with an actual person than just a computer. I mean, God forbid I'm only useful for what I can produce, and disposable when I reach my expiration date. I would like to see a twilight zone rendition of an AI dystopia where all the slow, ignorant and bothersome humans are replaced by lifeless AI
- baq 2y agoTime to re-read The Culture. Not everything has to end in a dystopia.
- hakanderyal 2y agoIt's not just about efficiency. I don't have the means to hire a junior right now, but 20$ is a no brainer.
- candiddevmike 2y agoManagement will be easier to replace than SWEs. I'm thinking there will come a time, similar to the show Mrs Davis, where AI will direct human efforts within organizations. AI will understand its limits and create tasks/requirements for human specialists to handle.
- al_borland 2y agoMy first thought with this is that AI would be directed to figure out exactly how little people are willing to work for, and how long, before they break. I hope I’m wrong, and it instead shows that more pay and fewer hours lead to a better economy, because people have money and time to spend it… and output isn’t impacted enough to matter.