That’s a great name! Better than my “Task-Enabled Autonomous Agent” (thanks GPT4 – who named it). Nickname: BabyAGI
AGI
-
Building Perpetual Agents: Autonomous AI Systems Design
By
–
Why do you think I’m building a perpetual agent 🙂
-
Cleaning Up Perpetual Agent Architecture
By
–
Cleaning up my perpetual agent 🙂 https://t.co/DsF3Q3LMLR
— Yohei (@yoheinakajima) 2 avril 2023Cleaning up my perpetual agent 🙂
-
AI System Tackles Global Warming with Expanding Reasoning
By
–
Tried the objective: fix global warming
— Yohei (@yoheinakajima) 2 avril 2023
Still really cool to see it perpetually expand its thinking on this one topic.https://t.co/rvrgwnAOMRTried the objective: fix global warming Still really cool to see it perpetually expand its thinking on this one topic.
-
AI Solves Global Warming in Record Time with Safety Concerns
By
–
Oh, that's a fun one. Didn't quite let it get to wiping out humans (also, I need to work on memory better to reduce deduping after some time).
— Yohei (@yoheinakajima) 2 avril 2023
Objective: fix global warming (1 min 43 min) pic.twitter.com/mm6bYGtsATOh, that's a fun one. Didn't quite let it get to wiping out humans (also, I need to work on memory better to reduce deduping after some time). Objective: fix global warming (1 min 43 min)
-
World Interaction as Key Learning and Evolution Method
By
–
More or less, but being able to interact w the world is probably the best way to learn and evolve
-
Simplifying AI Agent Architecture with OpenAI Text-Completion
By
–
Basically stripping it down to a barebones template only using OpenAI text-completion for all three task agents (execution, creation, prioritization). I think it makes sense to clean it up like this, and then add back on complexity – like Langchain, etc.
-
Simplifying AI Agent Architecture by Removing Langchain Framework
By
–
Taking a pause from adding complexity to the agent as it's harder to debug.
— Yohei (@yoheinakajima) 2 avril 2023
Took it an opposite direction and stripped out Langchain for now. It can't engage with APIs now, but can still perpetually build and execute tasks (just writing and "thinking"). pic.twitter.com/rJKDXZEchbTaking a pause from adding complexity to the agent as it's harder to debug. Took it an opposite direction and stripped out Langchain for now. It can't engage with APIs now, but can still perpetually build and execute tasks (just writing and "thinking").
-
Task Management AI Agent Execution and Reprioritization Issues
By
–
It then continued to:
– execute tasks
– create new tasks
– reprioritize tasks Task mgmt agent is a bit wonky, so need to fix that… btw, all text-davinci-003 cuz… that's what i know -
Projecting Our Values onto More Intelligent Systems
By
–
seems wrong to project our own values/frameworks on a being that is more intelligent than us we have zero understanding (and probably never will) of what a more intelligent system would want/desire otherwise we would be the more intelligent system