Submission declined on 26 November 2023 by TheChunky (talk). This submission is not adequately supported by reliable sources. Reliable sources are required so that information can be verified. If you need help with referencing, please see Referencing for beginners and Citing sources.
Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
This draft has not been edited in over six months and qualifies to be deleted per CSD G13. Declined by TheChunky 6 months ago. Last edited by TheChunky 6 months ago. Reviewer: Inform author.
|
AI agent (or Autonomous AI agent or Autonomous LLM agent) is autonomous program powered by artificial intelligence, demonstrating the ability to independently generate, prioritize, and execute tasks in pursuit of a given objective, iterating until the objective is achieved[1]. AI agents provide AI apps new capabilities through a combination of reasoning and planning, self-reflection, tool usage, and memory.[2]
Examples of popular AI agents are Auto-GPT, GPT-Engineer, Superagent, Aomni, or BabyAGI. These agents are designed to go beyond generating text and code; they function as versatile problem solvers. Autonomous agents can be applied across fields and serve variety of tasks, from managing a social media account, investing in the market, to coming up with the best children’s book.[3]
In a LLM-powered autonomous agent system, the LLM serves as the agent's brain, supported by three key components:
Subgoal and decomposition: The agent breaks down large tasks into smaller, manageable subgoals, enabling efficient handling of complex tasks.[4]
Reflection and refinement: The agent can do self-criticism and self-reflection on past actions, learn from mistakes and refine them for future steps, thereby improving the quality of final results.[5]
Short-term Memory: Utilizes in-context learning for task-specific information.
Long-term Memory: Retains and recalls information over extended periods using external storage.
Agents can use "tools" by calling external APIs - for example, it can browse the web, use apps, read and write files, make payments, and even control a user's laptop[6]. Learns to utilize external APIs for additional information, enhancing capabilities beyond pre-training limitations.
On March 30, 2023, Auto-GPTwas released by Toran Bruce Richards, the lead game developer at video game company Significant Gravitas Ltd. Auto-GPT is an open-source autonomous AI agent based on OpenAI’s API for GPT-4, the large language model released on March 14, 2023. Auto-GPT is among the first examples of an application using GPT-4 to perform autonomous tasks.
Since then, AI agents have experienced a boom during Summer 2023.
OpenAI, that indicated before that it intended to allow users to define their own customizable AI agents[7], launched "GPTs" in November 2023[8], and the Assistants API, a developer-facing part of the GPTs. OpenAI avoided the term “AI agent” and used “GPTs”, even though they follow the characteristics of agents.[9]
Anticipated within the next five years, the advent of AI agents promises a fundamental change[10]. Users will communicate with devices in everyday language, eliminating the need for multiple apps for distinct tasks. AI agents respond to natural language and execute various tasks based on a comprehensive understanding of the user, a concept that has evolved over nearly three decades and recently became practical due to AI advancements.
Today’s most common use cases are coding, personal daily tasks, or research. We may expect a further shift towards a vertical market, for example, one app with different underlying agents designed for code writing, code debugging, code migration, e-mail communication, calendar planning, and task management.[11]
Beyond transforming individual interactions, AI agents are predicted to revolutionize the software industry.[12] This shift is likened to the monumental change from typing commands to tapping on icons, signaling a significant computing paradigm shift. Andrej Karpathy envisions LLMs as kernels in a new operating system, indicating their growing importance[13].
Hundreds to thousands of people started to identify as AI Engineers[14]. There is a new ecosystem being formed around AI agents.[15]