As AI technology continues to advance, developers are working hard to create autonomous AI agents that allow for faster and more effective development. The rise of autonomous AI agents like Auto-GPT, BabyAGI and AgentGPT has been a major step in this race. This development has revolutionized the way artificial intelligence is created, making it easier to use and implement.
Auto-GPT was launched in late March by Toran Bruce Richards, under the name @significantgravitas, and it soon became the fastest-growing repository on GitHub in history. Auto-GPT is an ‘experimental open-source application’ connected to OpenAI’s GPT-4 through an API and running on Python. This AI tool does not require human involvement to prompt actions, making it autonomous, and has access to internet, long/short-term memory, and can string GPT calls together.
Shortly after, BabyAGI was launched by @yoheinakajima on March 29. It is a “task-driven autonomous agent” that is powered by GPT-4, Pinecone’s vector search, and LangChainAI’s framework, allowing it to autonomously create and perform tasks based on an objective.
Jay Scambler, a consultant and strategist, said the tools are game-changing and can create a team of AI agents without much effort. Not all share his enthusiasm though—Nvidia AI scientist Jim Fan warned that AutoGPT is a fun experiment, but it is far from becoming production-ready.
Experts points out that both Auto-GPT and BabyAGI are difficult to access by the average ChatGPT user, and, Nicola Bianzino, chief technology officer at EY, admits he does not completely understand all the variations that are out there in the wild. Although they bring a lot of potential, these autonomous agents also come with risks, such as errors and malicious use cases like cyberattacks and fraud.
HyperWrite unveiled an AI agent recently that can browse the web and interact with websites, and AgentGPT was launched on April 8 as an easier access for non-tech users. While some are claiming that this is just an example of “hustle bros” over-hyping the potential of these tools to make a profit, the truth is that the pace of AI development in this space is undeniably quick and worth monitoring.
Toran Bruce Richards, creator of Auto-GPT, is a developer and entrepreneur who is the co-founder of significantgravity.com and founder of decentralized networks and AI-based startups. Richards was formerly a professor of AI and robotics at Stanford University, and a director of research at OpenAI. He is a highly sought-after advisor and mentor, and the main focus of his work is to deploy AI to further our understanding of the universe.
EY (Ernst and Young), who mentioned in this article, is one of the world’s largest professional services firms providing assurance, tax, transaction and advisory services to businesses. It has more than 200,000 employees and operations in more than 150 countries globally with a revenue of US$36.4 billion in the 2017-18 financial year, making them one of the leading professional services firms in the world. EY focuses on all facets of business, transforming industry processes, and providing expertise on how to improve customer experience. It also employs thousands of professionals in areas like AI and blockchain.
