AutoGPT
In plain wordsA 2023 open-source program that gave GPT-4 a goal and let it run on its own — planning, searching the web and writing files in a loop — becoming the first viral 'autonomous AI agent'.
An open-source Python application that wrapped OpenAI's GPT-4 (or GPT-3.5) in an autonomous loop: given a name, a role and a few goals, the model repeatedly wrote out its next thought, reasoning, plan and self-criticism, chose a command — web search, browsing, reading and writing files, running code, starting helper instances — and received the result, with a vector database as memory beyond its context window. Its striking demonstrations and failures shaped public expectations of agentsAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.: runs often looped, lost track of progress, invented results or ran up API bills without finishing, exposing the gap between model fluency and reliable autonomy. The project later turned into a platform for building and running agent workflows.
Like many people in 2023, I watched AutoGPT burn through API credits while looping on the same web search.
Origin
Released on GitHub on 30 March 2023 by Toran Bruce Richards, founder of the games company Significant Gravitas, two weeks after GPT-4. It became GitHub's top trending repository within weeks and inspired many imitators, including BabyAGIBabyAGI, prop. n.: A tiny 2023 program that kept an AI working through an ever-changing to-do list — doing a task, adding new ones, reprioritizing — and helped start the craze for autonomous AI agents. (April 2023).
Usage note
AutoGPT predated reliable function callingfunction calling, n.: A feature of AI model services that lets the model reply with a neatly formatted request to run one of a program's functions, rather than with ordinary text; the program, not the model, then runs it. and parsed commands from free text, one reason for its brittleness. Its failures helped make stop conditionsstop condition, n.: The rule that tells an AI agent when to stop working — because the task is done, a time or cost limit has been reached, or something has gone wrong., human approval and narrowly scoped tasks standard agent design.
Related
- AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done.
- BabyAGIBabyAGI, prop. n.: A tiny 2023 program that kept an AI working through an ever-changing to-do list — doing a task, adding new ones, reprioritizing — and helped start the craze for autonomous AI agents.
- agent loopagent loop, n.: The repeating cycle at the heart of an AI agent: the model decides what to do next, software carries it out, the result is shown to the model, and the cycle repeats until the task is finished.
- stop conditionstop condition, n.: The rule that tells an AI agent when to stop working — because the task is done, a time or cost limit has been reached, or something has gone wrong.
- agentic AIagentic AI, n.: AI built to pursue goals on its own — planning, taking actions and adjusting along the way — instead of only responding to one prompt at a time. 'Agentic' describes how much of this independence a system has.
- agent frameworkagent framework, n.: A software library that gives developers ready-made parts for building AI agents — model connections, tool handling, memory, multi-agent coordination and logging — so they need not write everything from scratch.
Sources
- Wikipedia. AutoGPT. en.wikipedia.org/wiki/AutoGPT
Mentioned in
AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done., agent frameworkagent framework, n.: A software library that gives developers ready-made parts for building AI agents — model connections, tool handling, memory, multi-agent coordination and logging — so they need not write everything from scratch., agent loopagent loop, n.: The repeating cycle at the heart of an AI agent: the model decides what to do next, software carries it out, the result is shown to the model, and the cycle repeats until the task is finished., scaffoldingscaffolding, n.: The code, prompts and tools built around an AI model so it can carry out tasks it could not manage from a bare prompt — like the scaffolding that lets builders reach every part of a building., BabyAGIBabyAGI, prop. n.: A tiny 2023 program that kept an AI working through an ever-changing to-do list — doing a task, adding new ones, reprioritizing — and helped start the craze for autonomous AI agents., stop conditionstop condition, n.: The rule that tells an AI agent when to stop working — because the task is done, a time or cost limit has been reached, or something has gone wrong.