Amazon Launches AGI SF Lab to Develop Advanced AI Agents

Amazon opens a new laboratory to develop artificial intelligence agents - Softonic

Amazon has announced the creation of the Amazon AGI SF Lab, a new research and development laboratory located in San Francisco. This project will focus on designing fundamental capabilities for artificial intelligence agents capable of performing actions in both the digital and physical worlds. David Luan, co-founder of the startup Adept, will lead the laboratory, which will develop tools to manage complex workflows using computers, browsers, and other code interpreters. "Our work will be based on Amazon's broader AGI team," explained Luan and Pieter Abbeel, a robotics researcher who joined Amazon after an agreement with Covariant, in a statement. Abbeel will work closely with Luan on the initial development of AI agents capable of performing real actions, learning from human feedback, self-correcting, and understanding objectives. The lab will have Adept employees as its foundation, and Amazon plans to hire several dozen researchers specializing in areas such as quantitative finance, physics, and mathematics. Last June, Amazon reached an agreement with Adept to license its technology. This allowed Luan and part of his team to integrate into the e-commerce giant under the supervision of Rohit Prasad, former leader of Alexa, who now leads a general artificial intelligence team specializing in large language models. Founded two years ago, Adept emerged with the goal of developing AI models capable of performing actions in any software tool through natural language. According to Luan, their vision was to create a versatile "AI companion" that could handle various applications. This concept of agentic AI has been gaining great popularity, and a report from Emergen Research estimates that the sector will reach a value of 31 billion dollars by the end of the year. Although OpenAI, Google, and other major tech companies are working on similar developments, Amazon has shown great interest in AI over the past year. Recently, it announced conversational agents for Bedrock AI and added similar features to its Amazon Q Business platform. Additionally, Andy Jassy, CEO of Amazon, hinted that Alexa could evolve towards a more "agent" approach, combining responses with action capability.

Amazon forms a new AI agent-focused lab led by Adept co-founder | TechCrunch

Amazon says that it's establishing a new R&D lab in San Francisco, the Amazon AGI SF Lab, to focus on building "foundational" capabilities for AI agents. The Amazon AGI SF Lab, which will be led by David Luan, the co-founder of AI startup Adept, will seek to build agents that can "take actions in the digital and physical worlds," and "handle complex workflows" using computers, web browsers, and code interpreters. "Our work will build on that of Amazon's broader AGI team," reads a post jointly written by Luan and Pieter Abbeel, a robotics research lead who joined Amazon by way of the company's acquisition of Covariant. An Amazon spokesperson tells TechCrunch that Abbeel will be working "closely" with Luan and the AGI SF Lab going forward. "Our initial focus is on several key research bets that will enable AI agents to perform real-world actions, learn from human feedback, self-course-correct, and infer our goals," added Luan and Abbeel. The lab will be seeded by Adept employees, and Amazon says it's looking to hire a few "dozen" additional researchers in fields such as quantitative finance, physics, and math. In June, Adept, which is developing AI-powered agents to complete software-based tasks, agreed to license its tech to Amazon, and Luan and portions of Adept's team joined the e-commerce giant. Luan was working under -- and will continue to work under -- Rohit Prasad, the former head of Alexa who's leading an AGI team specializing in large language models. Amazon's quasi-acquisition of Adept resembled the deal Microsoft struck with AI company Inflection in May. Both have come under regulatory scrutiny as policymakers stateside and abroad seek to determine whether tech giants are smothering their AI rivals. Adept was founded two years ago with the goal of creating an AI model that can perform actions on any software tool using natural language. At a high level, the vision was to create an "AI teammate," of sorts, trained to use a wide variety of different software tools and APIs. Many others now share this vision. According to Emergen Research, "agentic" AI could be worth $31 billion as a sector by the end of the year. Eighty-two percent of orgs say that they plan to integrate AI agents within three years, per a Capgemini poll -- attracted by the possible efficiency boosts. In addition to startups like Orby, Emergence, and Rabbit, OpenAI and other major AI players are developing agent products to complete tasks largely autonomous. OpenAI rival Anthropic earlier this year released its take on the tech, while Google is reportedly working on AI agents that can make purchases, such as booking flights and hotels. Amazon has dabbled in the agent space, but has yet to make a serious play. In July, the company announced conversational agents for its Bedrock AI development platform, and just last week, it brought agents to its Amazon Q Business assistant platform for business customers and devs. Amazon CEO Andy Jassy has hinted at a more agentic Alexa, meanwhile -- one capable of not only responding to questions but taking actions.

Amazon Opens New AI Lab Focused on AI Agents

Google, Anthropic, and Microsoft are also working on AI agents Amazon announced setting up a new artificial intelligence (AI) laboratory on Monday. Dubbed Amazon AGI SF Lab, the new research lab will be based in San Francisco and focus on developing AI agents with real-world applications. The Seattle-based tech recently hired several top executives from the startup Adept AI Labs and will be seeding the new division with the new hires. Additionally, the Amazon AGI SF Lab will be headed by the former CEO and Co-founder of Adept, David Luam. In a blog post, the tech giant stated that the new research unit will be focusing on developing foundational capabilities for AI agents that can take action in both digital as well as physical worlds. Notably, AI agents can be understood as smaller and more efficient AI chatbots that can also execute actions by integrating with specialised hardware and software. Amazon highlighted that the team's responsibility will include building "practical AI" that can perform tasks for the company as well as its clients. As the division's name suggests, the team will also work closely with Amazon's artificial general intelligence (AGI) team, which recently introduced the Nova family of foundational models. "Our initial focus is on several key research bets that will enable AI agents to perform real-world actions, learn from human feedback, self-course-correct, and infer our goals," the company stated in the blog post. Amazon also highlighted that the team will combine large language models (LLMs) with reinforcement learning (RL) to solve reasoning and planning, learned world models, and generalising agents to physical environments. Amazon also announced that it is looking to hire "a few dozen" people for the Amazon AGI SF Lab. The company is looking for AI experts who have trained state-of-the-art (SOTA) models as well as candidates from other fields such as physics, mathematics, quantitative finance, and others. The blog post also added that experience level is not a criterion for hiring. Last week, Amazon Web Services (AWS) introduced the Nova family of artificial intelligence (AI) models at its re:Invent conference, with three text-based models, one image generation model, and a video generation model. All of these models are available for the company's enterprise clients and can be availed from the Amazon Bedrock platform.

The Future of AI Agents Belongs to Amazon, Thanks to AGI SF Lab

Amazon has announced a new dedicated team, which forms what is called the 'Amazon AGI SF Lab', based in San Francisco, United States. The new team will work on developing highly capable AI agents that can 'take actions in digital and physical worlds'. "Our work will build on that of Amazon's broader AGI team, which recently introduced Amazon Nova, a new generation of state-of-the-art foundation models (FMs)," said Amazon in the official announcement. Interestingly, this new lab will be 'seeded' by the team hired by Adept, an AI startup whose founders were lured into Amazon earlier this year. Amazon also said that they're keeping this team 'lean' to ensure each person receives a maximised amount of computing. "Our initial focus is on several key research bets that will enable AI agents to perform real-world actions, learn from human feedback, self-course-correct, and infer our goals," read the announcement. Amazon also shared their excitement in 'combining' large language models with reinforcement learning to solve 'solve reasoning and planning, learned world models, and generalising agents to physical environments'. "We're entering an exciting new era where AI agents are the next playing field; the right research bets can reinvent what's possible with AI," added Amazon. Amazon also said they're on the lookout for passionate and talented people to join their team, not just AI experts, 'but also candidates from other disciplines who will bring fresh thinking to the field, such as physics, math, or quantitative finance, regardless of experience level'. Only a few days ago, Amazon announced a new family of foundational models, the Nova. At the AWS re:Invent conference in Las Vegas, Amazon announced the Nova Micro, Lite, Pro and Premier, available exclusively on Amazon Bedrock. The flagship Nova Pro modal offers performance parity with other industry-leading models like Claude 3.5 Sonnet and OpenAI's GPT-4o. "They're really cost-effective. They're about 75% less expensive than the other leading models in Bedrock," said Amazon's CEO Andy Jassy. The Nova models also possess multimodal read and write capabilities and can process text, image, and video inputs. "The models are integrated with the distillation feature, which allows you to infuse the intelligence of larger models into smaller models that are more cost-effective and offer lower latency," Jassy added. Recently, Jeff Bezos announced his active return to Amazon and expressed his commitment to steering the company toward groundbreaking innovations. "My heart is in Amazon, my curiosity is Amazon, my fears are there, my love is there, I'm never going to forget about Amazon. I'll always be there to help," Bezos stated in a recent interview. His involvement is expected to accelerate the AGI SF Lab's ambitions, reinforcing Amazon's position at the forefront of AI advancements. "We are literally working on a thousand applications internally. You have to remember modern AI is a horizontal enabling layer. It can be used to improve everything; it will be in everything. This is most like electricity," said Bezos. Bezos also emphasised the importance of tackling large-scale challenges with a clear focus, stating, "Big leaders only have to do a few things. Big leaders have to identify the big ideas, they have to enforce tough execution, right against those big ideas, and they need to grow the next generation of leaders."