I have been on a deep dive for the last six months, using AI to do software development for a property management business. Today, a teaser: how I used it to write code while cooking bacon at home
Brilliant! I liked this: "Instead of 4-6 hours of work for expensive programmers, I got working code to resolve these bounced email message in under a minute while cooking bacon." I don't have a Monopoly board full of real estate properties but I would be calling you today if I did.
Great explanation of the process of creating an expert system for your business management needs.
I finally got a chance to query an AI (Perplexity AI) on the benefits and risks of a laparoscopic robotic surgery under general anesthesia inguinal hernia repair that the doctor was scheduling me for. It gave good answers and follow-up questions and information I could check out. It also listed the sources it used. Upshot: at my age the risks were significant and I was better off doing traditional surgery under local anesthesia. I went to the source material and was able to see the accuracy of what the AI concluded and why. I am impressed with this beneficial use of AI for us average folk.
Your experience mirrors what I have been doing for the past year or so. It's quite a productivity boost!
I think that in the hands of an experienced software developer or engineer, this is a remarkable tool. In the hands of someone who is lacking experience and who isn't able to properly design, direct, audit and evaluate the code that is produced by the LLM AI, it could end up producing code that has security holes or hidden bugs.
I agree with you that this is a game-changer and that there is no going back. Also, that it is unnerving as there is machine reasoning that is going on here.
One more comment: I find that the quality and correctness of the code produced by the LLM A.I. is remarkably good when I ask it to program in a language that has a large user-base and a lot of example code available for it to have been trained on. However, when I ask it to code in a somewhat obscure language, the code that is produced is not nearly as good.
You’ve just described how improved technology (AI) can be used for GOOD!!
If there weren’t evil people in the world trying to cull the herd and completely control the rest, I wouldn’t worry about AI (especially a reasoning AI) being used for bad.🤔
God bless you, Cognitive Carbon, as you obviously are one of those God has gifted for this time.🙏
Brilliant! I liked this: "Instead of 4-6 hours of work for expensive programmers, I got working code to resolve these bounced email message in under a minute while cooking bacon." I don't have a Monopoly board full of real estate properties but I would be calling you today if I did.
Great explanation of the process of creating an expert system for your business management needs.
I finally got a chance to query an AI (Perplexity AI) on the benefits and risks of a laparoscopic robotic surgery under general anesthesia inguinal hernia repair that the doctor was scheduling me for. It gave good answers and follow-up questions and information I could check out. It also listed the sources it used. Upshot: at my age the risks were significant and I was better off doing traditional surgery under local anesthesia. I went to the source material and was able to see the accuracy of what the AI concluded and why. I am impressed with this beneficial use of AI for us average folk.
Your experience mirrors what I have been doing for the past year or so. It's quite a productivity boost!
I think that in the hands of an experienced software developer or engineer, this is a remarkable tool. In the hands of someone who is lacking experience and who isn't able to properly design, direct, audit and evaluate the code that is produced by the LLM AI, it could end up producing code that has security holes or hidden bugs.
I agree with you that this is a game-changer and that there is no going back. Also, that it is unnerving as there is machine reasoning that is going on here.
One more comment: I find that the quality and correctness of the code produced by the LLM A.I. is remarkably good when I ask it to program in a language that has a large user-base and a lot of example code available for it to have been trained on. However, when I ask it to code in a somewhat obscure language, the code that is produced is not nearly as good.
You’ve just described how improved technology (AI) can be used for GOOD!!
If there weren’t evil people in the world trying to cull the herd and completely control the rest, I wouldn’t worry about AI (especially a reasoning AI) being used for bad.🤔
God bless you, Cognitive Carbon, as you obviously are one of those God has gifted for this time.🙏
I agree with your statement SO MUCH!