Agentic AI

David Silver and Richard Sutton, two renowned AI scientists, argue in a new paper that artificial intelligence is about to enter a new phase, the “Era of Experience.” This is where AI systems rely increasingly less on human-provided data and improve themselves by gathering data from and interacting with the world.

So-called agentic artificial intelligence doesn’t require a human to tell it what to do. At least, that’s the idea.

AI co-scientist is a multi-agent AI system built with Gemini 2.0 as a virtual scientific collaborator to help scientists generate novel hypotheses and research proposals, and to accelerate the clock speed of scientific and biomedical discoveries.

"Agentic AI, an emerging paradigm in artificial intelligence, refers to autonomous systems designed to pursue complex goals with minimal human intervention. Unlike traditional AI, which depends on structured instructions and close oversight, Agentic AI demonstrates adaptability, advanced decision-making capabilities and self-sufficiency, enabling it to operate dynamically in evolving environments. This survey thoroughly explores the foundational concepts, unique characteristics, and core methodologies driving the development of Agentic AI."

“Agentic” is the latest buzzword in AI, and in this case, it’s appropriate to believe the buzz. Agentic AI brings together the versatility and flexibility of large language models (LLMs) and the precision of traditional programming.

How does agentic AI work? What can it do for your organization? What security issues should be on your radar screen? Catch up on key information from MIT SMR experts.

The future of AI assistance lies in task delegation, says the author, where users assign responsibilities to AI systems, enabling them to autonomously handle tasks and deliver results. And that future, for everyday users, is here: In October 2024, Anthropic, which researches and develops AI to encourage safe and reliable deployment of the technology, introduced a new feature for their AI assistant, Claude, that allows the bot to perform tasks directly on a user’s computer.

To perform complex tasks, like booking a flight, AI agents will need permissions to work on behalf of a person.

Last year, big tech couldn’t stop talking about how AI “agents” would be the next big thing in 2025. It hasn’t quite turned out that way.

This study attempts to rethink evaluations and benchmarks to build AI agents that are useful in the real world

The new “Computer Use” feature allows AI to perform tasks that were previously handled only by human workers, such as opening applications, interacting with interfaces, and filling out forms.

ChatGPT agent represents a fundamental shift. It is not trained on the process of doing work; instead, OpenAI used reinforcement learning to train their AI on the actual final outcomes.

Can AI agents be trusted to act in our best interests? Will they work exclusively for us, or will their loyalty be split between users, developers, advertisers, and service providers? And how will we know?

There is enormous potential for this technology, but only if we deploy it responsibly.

The first narrow agents, Open AI's Deep Research and Google's Deep Research, are now available.

The foundation of current AI literacy, focused primarily on crafting effective prompts and evaluating text responses, is increasingly inadequate for sophisticated agentic systems.

This collection of AI agents is arranged into categories that make it easy to find particular agents for particular tasks.

FutureHouse helps accelerate scientific discovery with specialized agents that search and synthesize scientific literature more effectively than PhD researchers.

This site provides agents for conducting agentic deep research as well as agentic data searching.

Anthropic offers the new function that enables its Sonnet version to control your computer. “Claude 3.5 Sonnet is the first frontier AI model to offer computer use in public beta.” This is just the beginning of the next phase of GenAI. Over the coming weeks and months, we will see many platforms offer this capability to users.

This survey bridges the existing gap by presenting an exhaustive blueprint for scientific agents, spanning systematic construction methodologies, targeted capability enhancement, and rigorous evaluations.

As AI agents evolve and risk pathways emerge, governance frameworks must keep pace. Our analysis identifies open questions that present clear opportunities for the European Union to act on, while other jurisdictions should also take note of the EU’s approach.

A new report by FutureSearch, titled Deep Research Bench (DRB): Evaluating Web Research Agents, offers the most rigorous evaluation to date—and the results reveal both impressive capabilities and critical shortcomings.

What makes "agentic workflows" possible for everyone, not just programmers, is no-code and low-code automation platforms. These are software tools that let you build automations visually, often without writing a single line of code. Think of them like digital LEGOs for your workflow.

Manus is a general AI agent that bridges minds and actions: it doesn't just think, it delivers results. Manus excels at various tasks in work and life, getting everything done while you rest.

Manus Wide Research lets you research hundreds of items at once using multiple AI agents working together in parallel on cloud infrastructure.

The technique could make AI systems better at complex tasks that involve variability.

AI agents are autonomous systems capable of sensing, learning and acting upon their environments. This white paper explores their development and looks at how they are linked to recent advances in large language and multimodal models. It highlights how AI agents can enhance efficiency across sectors including healthcare, education and finance.

Agentic AI is poised to revolutionize industries, handling complex tasks with human-like decision-making—but with great power comes great risk.

OpenAI has revealed another new agentic feature for ChatGPT called deep research, which it says can operate autonomously to “plan and execute a multi-step trajectory to find the data it needs, backtracking and reacting to real-time information where necessary."

Hugging Face has launched Open Computer Agent, a free cloud-based AI that performs tasks on a virtual Linux machine.

Leading Stanford faculty offer their expectations for agentic AI, and artificial intelligence more generally in the new year.

AI agents can accomplish far more steps than they could before and can use tools (which basically include anything your computer can do) without substantial human intervention.

From OpenAI’s Deep Research and Google’s Gemini counterpart to academic-focused players like Elicit, Undermind.ai and SciSpace, these agentic systems promise comprehensive reports and literature reviews—albeit with a much slower response.

In collaboration with Capgemini, the World Economic Forum has published a new white paper, Navigating the AI Frontier: A Primer on the Evolution and Impact of AI Agents, which explores the capabilities and implications of AI agents, to provide stakeholders with a better understanding of how these systems can drive meaningful progress across sectors.

The authors take the position that small language models (SLMs) are sufficiently powerful, inherently more suitable, and necessarily more economical for many invocations in agentic systems, and are therefore the future of agentic AI

Build advanced, multi-task AI Agents—no coding required. You can build anything from a simple chatbot to a sophisticated AI Agent that can perform hundreds of distinct tasks. Free, open source from MIT.

Researchers from Princeton, Tsinghua, and other schools published the first comprehensive survey of self-evolving agents, establishing a unified framework for AI systems that continuously adapt and improve toward Artificial Super Intelligence.

Recently, significant progress in large language models has fueled new optimism about the prospect of building sophisticated AI agents. This CSET-led workshop report synthesizes findings from a May 2024 workshop on this topic, including what constitutes an AI agent, how the technology is improving, what risks agents exacerbate, and intervention points that could help.

This blog post looks into the top 20 use cases for agentic AI and analyzes real-life examples to illustrate its impact.

AI could soon not only mimic our personality, but go out and act on our behalf. There are some things we need to sort out before then.

This article explains how AI agents work, why people are jazzed about them, and what risks they hold.

The agentic AI prize could be great, with the promise of greater productivity, innovation and insights for the human workforce. But so, too, are the risks: the potential for bias, mistakes, and inappropriate use.

Agentic AI refers to advanced artificial intelligence systems that have the ability to make decisions on their own in complex environments. They go beyond rule-based or generative models and demonstrate a high degree of autonomy and adaptability.
Agentic AI can set objectives, devise strategies, and execute multistep tasks with minimal human supervision.

The new Claude computer use model announced today shows us a hint of what an agent means. It is capable of some planning, it has the ability to use a computer by looking at a screen (through taking a screenshot) and interacting with it (by moving a virtual mouse and typing), It is a good preview of an important part of what agents can do.

Beyond OCR: Intelligent Document Understanding with Visual Context. 
Convert decades of archived documents into LLM-ready data in hours rather than weeks.

This voice-enabled AI assistant autonomously browses the web, writes code, and plans tasks while keeping all data on your device. Tailored for local reasoning models, it runs entirely on your hardware, ensuring complete privacy and zero cloud dependency.

Built to address the challenges scientists face in drilling down into structured datasets, Asta DataVoyager delivers data-driven discovery and analysis capabilities that allow you to make queries about structured files in plain language and get clearly cited, explainable answers with copyable code, clear visuals, and a concise, well-supported summary.

Chatbase is a platform for building & deploying AI support agents for your organization.

ChatGPT now thinks and acts, proactively choosing from a toolbox of agentic skills to complete tasks for you using its own computer.

Claude operates agentically, conducting multiple searches that build on each other while determining exactly what to investigate next. It explores different angles of your question automatically and works through open questions systematically.

Perplexity's Comet Browser is an AI assistant that allows you to use spoken commands that can trigger clicks, form filling, and navigation inside sites, not just tab management.

CRESt can help to craft and run experiments by retrieving and analyzing data, turning equipment on and off using digital switches, powering robotic arms, documenting findings and alerting by email when issues arise or protocols end.

Get up to speed on just about anything with Deep Research, an agentic feature in Gemini that can automatically browse up to hundreds of websites on your behalf, think through its findings, and create insightful multi-page, reports that you can turn into engaging podcast-style conversations.

Super Agent is a no-code, customizable AI agent that helps users automate real-world tasks like making a phone call, generating a presentation, or turning a recipe into a short video.

Create agents with advanced memory that can learn and improve over time. Open source AI, built for developers.

Build AI agents in minutes to automate workflows, save time, and allow you to focus on other things.

Researcher helps you tackle complex, multi-step research at work—delivering insights with greater quality and accuracy than previously possible. You can use. Built on OpenAI’s o3-mini reasoning model and optimized to do advanced data analysis at work, Analyst uses chain-of-thought reasoning to progress through problems iteratively, taking as many steps as necessary to refine its reasoning and provide a high-quality answer that mirrors human analytical thinking.

NanoBrowser is an open-source Chrome extension that lets you extract data, fill forms, and search and retrieve items across the web that meet specific search criteria.

An agent that uses reasoning to synthesize large amounts of online information and complete multi-step research tasks for you.

A research prototype exploring the future of human-agent interaction, starting with browsers.

Create your own superhuman teammates in minutes. Equip your Agents with organization knowledge and have them do work across 7,000+ apps — on command and while you sleep.

Agents Towards Production is your go-to resource for building GenAI agents that scale - from zero to production. Whether you're just starting out or refining your deployment stack, this repository gives you the tools, patterns, and code examples to do it right.

In this tutorial you will build an agent that can interact with a search engine. You will be able to ask this agent questions, watch it call the search tool, and have conversations with it.

Anthropic's “Building Better Agents” explains complex concepts simply. Users have reported using their framework to build a research assistant that actually cites sources correctly—unlike the one that confidently attributed Shakespeare quotes to Taylor Swift.

In this course taught by Andrew Ng, you’ll gain a fundamental understanding and practical knowledge to develop production-ready agentic applications, from design patterns to deployment and evaluation.

This free course features eleven lessons teaching everything you need to know to start building AI Agents.

OpenAI's “A Practical Guide to Building Agents” walks you through creating AI systems that take meaningful actions.