Create an AI agent that works from a set of goals:
To get good output from ChatGPT or another LLM, you usually have to feed it several prompts. But what if you could just give your AI bot a set of fairly broad goals at the start of a session and then sit back while it generates its own set of tasks to fulfill those goals? That's the idea behind Auto-GPT, a new open-source tool that uses the OpenAI API (same LLM as ChatGPT) to prompt itself, based on your initial input.
We've already seen a number of Twitter users talk about how they are using Auto-GPT for everything from creating marketing plans to analyzing market data for investments to preparing topics for a podcast. Based on our hands-on experience, we can't say that it always works well (we asked it to write a Windows 11 how-to and the result was awful), but it's early days and some tasks may work better than others.
If you want to try Auto-GPT on your computer, it is easy to install, and while there are a few sticky points in the process, we've found ways to work with them, or around them to write this condensed guide on how to create your own Auto-GPT AI to help you in your goals.
[...] You may or may not need to add payment information to your OpenAI account. By default, the system will give you a certain amount of free credits. In Editor-in-Chief Avram Piltch's case, it was $18 worth of free credit that he was able to use without entering any payment methods. You may not get as much free credit or may need to add a payment method to your OpenAI account to proceed.
The article has a step-by-step guide for getting up and running on a Windows machine. If you choose to add a payment method, make sure put a limit on how much money it can charge.
Let the chaos begin!
(Score: 1) by khallow on Saturday April 22, @04:21AM
(Score: 5, Interesting) by Rosco P. Coltrane on Saturday April 22, @09:01AM (5 children)
Turn off the computer and go socialize while everybody isn't out of a job yet - and still mostly unaware of what's coming to them - and society is still functioning normally. It's nice to taste normality while it lasts.
Ever since I had the brutal realization of the devastation that AI will soon bring about, when my company started to cancel new hires directly because of AI, I haven't put in a single hour of overtime. In fact, I've been flusing the overtime I already have on the book, to spend as much time as possible with my friends and family and enjoy my life before it - and everybody else's - gets totally upended.
(Score: 1, Interesting) by Anonymous Coward on Saturday April 22, @09:59AM (1 child)
Excellent plan!
Without realizing it last night, I've also started a similar weekend -- went to the local History Museum (where I haven't been for ages) to hear an in-person lecture, with other people(gasp) in the same room! The lecturer wrote a book recently and I bought a copy after his nice talk, had him sign it.
It's pouring rain here, so we're not going to be able to do much spring cleanup in the yard, I think I'm going to settle in with my new book (paper, not e-book).
(Score: 2) by hendrikboom on Sunday April 23, @08:20PM
It's hard, but not impossible, for an author to sign an ebook.
(Score: 5, Insightful) by Gaaark on Saturday April 22, @12:11PM
Yeah: hoping my 'planned retirement' doesn't come early.
I want to spend more time in my garden, but hoping it won't be because i have no job and need the garden to eat.
I welcome my AI overlord: maybe the sh*t it produces now will provide me with enough manure to keep me in asparagus and horseradish for years. Maybe I'll have enough to give to family as well: they too may need it to survive.
Heil AI!
--- Please remind me if I haven't been civil to you: I'm channeling MDC. ---Gaaark 2.0 ---
(Score: 0) by Anonymous Coward on Saturday April 22, @08:07PM
Isn't it more of the same? Quality of work degraded until cheap, replaceable commodity hires ("labor") can do it. The cost of mediocre just got lowered to free for machine-produced garble. The work may be correct, it may not be - caveat emptor. What could be worse than the present trend toward corporate authoritarian dystopia anyway? Bring on the robot overlords.
(Score: 2) by richtopia on Monday April 24, @02:48PM
Last weekend I installed bees in my beehive - it was very rewarding.
Now, I don't go out every day, so in my freetime I am exploring these AI agents. I doubt my job will be jeopardized by an AI soon, but I suspect there is a large opportunity to learn the tools to leverage LLM and either become a better engineer or even segue into a new career. This transition is moving faster than any other technology deployment I've seen, and I suspect the early movers will get rewarded.
(Score: 2, Insightful) by Anonymous Coward on Saturday April 22, @07:59PM (3 children)
...but who the fuck is going to read it?
(Score: 4, Insightful) by Ox0000 on Saturday April 22, @11:33PM (1 child)
This is exactly the question I asked myself as well:
If a human didn't bother writing the content, what's the point for another human to read it?
If someone else is sending me e-mails or other correspondence that they didn't even bother writing, what's the point in me reading it? That feels like a waste of my time. If I wanted AI drivel, I would have asked the AI myself.
(Score: 3, Insightful) by deimtee on Sunday April 23, @02:36AM
Obviously the next step is to have AI read the drivel that is sent to you. If you have a big enough cloud account, you too can have 17 trillion followers on Twatter.
No problem is insoluble, but at Ksp = 2.943×10−25 Mercury Sulphide comes close.
(Score: 0) by Anonymous Coward on Sunday April 23, @02:03PM
We will all have an AI agent on our email to detect and automatically flag to purge spam. Of course the other side knows this and will improve their methods on sending emails that look like humans sent them. Add a speeling mistake here, a misquoted proverb there, and eventually we will run out of energy as the war escalates to be point of needing to be powered by nuclear reactors just to read email.
The future is bright.
(Score: 2) by gnuman on Sunday April 23, @10:49AM
It's average, at best. It's like having argument on stackoverflow without knowing anything beyond it.