I used to use ChatGPT and Copilot, but then I stopped because they sucked too bad to justify the price. I spun up an Ollama instance on my server but I almost never use that either because it sucks really bad. I integrated it into my Home Assistant instance and it sucks so bad that when I ask it to simply turn off a light, it only succeeds about 1 in 5 times.
It might not even be a prompt problem, but an overkill to a simple job. Sounds like the breaking point might be voice recognition - which I really thought would be much further along by now with all the other advancements. It seems figuring out what unique voices are saying is one of the hardest things to do well, and then add in background noises. Probably why the drive-thru AI attempts were terrible, working that job as a human can be difficult.
I used to use ChatGPT and Copilot, but then I stopped because they sucked too bad to justify the price. I spun up an Ollama instance on my server but I almost never use that either because it sucks really bad. I integrated it into my Home Assistant instance and it sucks so bad that when I ask it to simply turn off a light, it only succeeds about 1 in 5 times.
‘You’re just prompting it wrong.’
-The people who say LLMs have achieved AGI
It might not even be a prompt problem, but an overkill to a simple job. Sounds like the breaking point might be voice recognition - which I really thought would be much further along by now with all the other advancements. It seems figuring out what unique voices are saying is one of the hardest things to do well, and then add in background noises. Probably why the drive-thru AI attempts were terrible, working that job as a human can be difficult.