In the context of generative artificial intelligence, AI agents (also referred to as compound AI systems or agentic AI) are a class of intelligent agents that can pursue goals, use tools, and take actions with varying degrees of autonomy. In practice, they usually operate within human-defined objectives, constraints, and available tools. == Overview == AI agents possess several key attributes, including goal-directed behavior, natural language interfaces, the capacity to use external tools, and the ability to perform multi-step tasks. Their control flow is frequently driven by large language models (LLMs). Agent systems may also include memory components, planning logic, tool interfaces, and orchestration software for coordinating agent components. AI agents do not have a standard definition. NIST describes agentic AI as an emerging area requiring standards for secure operation, interoperability, and reliable interaction with external systems. A common application of AI agents is task automation: for example, booking travel plans based on a user's prompted request. Companies such as Google, Microsoft and Amazon Web Services have offered platforms for deploying pre-built AI agents. Several protocols have been proposed for standardizing inter-agent communication, with examples including the Model Context Protocol, Gibberlink, and many others. Some of these protocols are also used for connecting agents to external applications. In December 2025, Linux Foundation announced the formation of the Agentic AI Foundation (AAIF), with the goal of ensuring agentic AI evolves transparently and collaboratively. == History == AI agents have been traced back to research from the 1990s, with Harvard professor Milind Tambe noting that the definition of an AI agent was not clear at the time. Researcher Andrew Ng has been credited with spreading the term "agentic" to a wider audience in 2024. == Training and testing == Researchers have attempted to build world models and reinforcement learning environments to train or evaluate AI agents. For example, video games such as Minecraft and No Man's Sky as well as replicas of company websites, have also been used for training such agents. == Autonomous capabilities == The Financial Times compared the autonomy of AI agents to the SAE classification of self-driving cars, likening most applications to level 2 or level 3, with some achieving level 4 in highly specialized circumstances, and level 5 being theoretical. == Cognitive architecture == The following are some internal design options for reasoning within an agent: Retrieval-augmented generation ReAct (Reason + Act) pattern is an iterative process in which an AI agent alternates between reasoning and taking actions, receives observations from the environment or external tools, and integrates these observations into subsequent reasoning steps. Reflexion, which uses an LLM to create feedback on the agent's plan of action and stores that feedback in a memory cache. A tool/agent registry, for organizing software functions or other agents that the agent can use. One-shot model querying, which queries the model once to create the plan of action. === Reference architecture === Ken Huang proposed an AI agent reference architecture, which consists of seven interconnected layers, with each layer building on the functionality of the layers beneath it: Layer 1: Foundation models - provide the core AI engines to power agent capabilities. Layer 2: Data operations - manage the complex data infrastructure required for AI agent operations, including Vector database, data loaders, RAG. Layer 3: Agent frameworks - sophisticated software and tools that simplify the development and management of the AI agents. Layer 4: Deployment and infrastructure - provide the robust technical foundation for running AI agents. Layer 5: Evaluation and observability - focus on assessing the safety and performance of AI agents. Layer 6: Security and compliance - a crucial protective framework ensuring AI agents operate safely, securely, and conform to regulatory boundaries. At this layer security and compliance features embedded into all the AI agent stack layers are integrated together. Layer 7: Agent ecosystem - represents the AI agents' interface with real-world applications and users. == Orchestration patterns == To execute complex tasks, autonomous agents are often integrated with other agents or specialized tools. These configurations, known as orchestration patterns or workflows, include the following: Prompt chaining: A sequence where the output of one step serves as the input for the next. Routing: The classification of an input to direct it to a specialized downstream task or tool. Parallelization: The simultaneous execution of multiple tasks. Sequential processing: A fixed, linear progression of tasks through a predefined pipeline. Planner-critic: An iterative pattern where one agent generates a proposal and another evaluates it to provide feedback for refinement. == Multimodal AI agents == In addition to large language models (LLMs), vision-language models (VLMs) and multimodal foundation models can be used as the basis for agents. In September 2024, Allen Institute for AI released an open-source vision-language model. Nvidia released a framework for developers to use VLMs, LLMs and retrieval-augmented generation for building AI agents that can analyze images and videos, including video search and video summarization. Microsoft released a multimodal agent model – trained on images, video, software user interface interactions, and robotics data – that the company claimed can manipulate software and robots. == Applications == As of April 2025, per the Associated Press, there are few real-world applications of AI agents. As of June 2025, per Fortune, many companies are primarily experimenting with AI agents. The Information divided AI agents into seven archetypes: business-task agents, for acting within enterprise software; conversational agents, which act as chatbots for customer support; research agents, for querying and analyzing information (such as OpenAI Deep Research); analytics agents, for analyzing data to create reports; software developer or coding agents (such as Cursor); domain-specific agents, which include specific subject matter knowledge; and web browser agents (such as OpenAI Operator). By mid-2025, AI agents have been used in video game development, gambling (including sports betting), cryptocurrency wallets (including cryptocurrency trading and meme coins) and social media. In August 2025, New York Magazine described software development as the most definitive use case of AI agents. Likewise, by October 2025, noting a decline in expectations, The Information noted AI coding agents and customer support as the primary use cases by businesses. In November 2025, The Wall Street Journal reported that few companies that deployed AI agents have received a return on investment. === Applications in government === Several government bodies in the United States and United Kingdom have deployed or announced the deployment of agents, at the local and national level. The city of Kyle, Texas deployed an AI agent from Salesforce in March 2025 for 311 customer service. In November 2025, the Internal Revenue Service stated that it would use Agentforce, AI agents from Salesforce, for the Office of Chief Counsel, Taxpayer Advocate Services and the Office of Appeals. That same month, Staffordshire Police announced that they would trial Agentforce agents for handling non-emergency 101 calls in the United Kingdom starting in 2026. In December 2025, the Department of Neighborhoods in Detroit, Michigan, in partnership with a local business, deployed a pilot project in two Detroit districts for an AI agent to be used for customer service calls. In February 2025, Thomas Shedd, the director of the Technology Transformation Services, proposed using AI coding agents across the United States federal government. A recruiter for the Department of Government Efficiency proposed in April 2025 to use AI agents to automate the work of about 70,000 United States federal government employees, as part of a startup with funding from OpenAI and a partnership agreement with Palantir. This proposal was criticized by experts for its impracticality, if not impossibility, and the lack of corresponding widespread adoption by businesses. In December 2025, the Food and Drug Administration announced that it would offer "agentic AI capabilities" to its staff for "meeting management, pre-market reviews, review validation, post-market surveillance, inspections and compliance and administrative functions." That same month, the United States Department of Defense launched GenAI.mil, an internal platform for American military personnel to use generative AI-based applications based on Google Gemini, including "intelligent agentic workflows". Defense Secretary Pete Hegseth listed applications such as "[conducting] deep r
Vicarious (company)
Vicarious was an artificial intelligence company based in the San Francisco Bay Area, California. They use the theorized computational principles of the brain to attempt to build software that can think and learn like a human. Vicarious describes its technology as "a turnkey robotics solution integrator using artificial intelligence to automate tasks too complex and versatile for traditional automations". Alphabet Inc acquired the company in 2022 for an undisclosed amount. == Founders == The company was founded in 2010 by D. Scott Phoenix and Dileep George. Before co-founding Vicarious, Phoenix was Entrepreneur in Residence at Founders Fund and CEO of Frogmetrics, a touchscreen analytics company he co-founded through the Y Combinator incubator program. Previously, George was Chief Technology Officer at Numenta, a company he co-founded with Jeff Hawkins and Donna Dubinsky while completing his PhD at Stanford University. == Funding == The company launched in February 2011 with funding from Founders Fund, Dustin Moskovitz, Adam D’Angelo (former Facebook CTO and co-founder of Quora), Felicis Ventures, and Palantir co-founder Joe Lonsdale. In August 2012, in its Series A round of funding, it raised an additional $15 million. The round was led by Good Ventures; Founders Fund, Open Field Capital and Zarco Investment Group also participated. The company received $40 million in its Series B round of funding. The round was led by individuals including Mark Zuckerberg, Elon Musk, and others. An additional undisclosed amount was later contributed by Amazon.com CEO Jeff Bezos, Yahoo! co-founder Jerry Yang, Skype co-founder Janus Friis and Salesforce.com CEO Marc Benioff. == Recursive Cortical Network == Vicarious is developing machine learning software based on the computational principles of the human brain. One such software is a vision system known as the Recursive Cortical Network (RCN), it is a generative graphical visual perception system that interprets the contents of photographs and videos in a manner similar to humans. The system is powered by a balanced approach that takes sensory data, mathematics, and biological plausibility into consideration. On October 22, 2013, beating CAPTCHA, Vicarious announced its model was reliably able to solve modern CAPTCHAs, with character recognition rates of 90% or better when trained on one style. However, Luis von Ahn, a pioneer of early CAPTCHA and founder of reCAPTCHA, expressed skepticism, stating: "It's hard for me to be impressed since I see these every few months." He pointed out that 50 similar claims to that of Vicarious had been made since 2003. Vicarious later published their findings in peer-reviewed journal Science. Vicarious has indicated that its AI was not specifically designed to complete CAPTCHAs and its success at the task is a product of its advanced vision system. Because Vicarious's algorithms are based on insights from the human brain, it is also able to recognize photographs, videos, and other visual data.
Reification (knowledge representation)
Reification in knowledge representation is the process of turning a predicate or statement into an addressable object. Reification allows the representation of assertions so that they can be referred to or qualified by other assertions, i.e., meta-knowledge. The message "John is six feet tall" is an assertion involving truth that commits the speaker to its factuality, whereas the reified statement "Mary reports that John is six feet tall" defers such commitment to Mary. In this way, the statements can be incompatible without creating contradictions in reasoning. For example, the statements "John is six feet tall" and "John is five feet tall" are mutually exclusive (and thus incompatible), but the statements "Mary reports that John is six feet tall" and "Paul reports that John is five feet tall" are not incompatible, as they are both governed by a conclusive rationale that either Mary or Paul is (or both are), in fact, incorrect. In linguistics, reporting, telling, and saying are recognized as verbal processes that project a wording (or locution). If a person says that "Paul told x" and "Mary told y", this person stated only that the telling took place. In this case, the person who made these two statements did not represent a person inconsistently. In addition, if two people are talking to each other, let's say Paul and Mary, and Paul tells Mary "John is five feet tall" and Mary rejects Paul's statement by saying "No, he is actually six feet tall", the socially constructed model of John does not become inconsistent. The reason for that is that statements are to be understood as an attempt to convince the addressee of something (Austin's How to do things with words), alternatively as a request to add some attribute to the model of Paul. The response to a statement can be an acknowledgement, in which case the model is changed, or it can be a statement rejection, in which case the model does not get changed. Finally, the example above for which John is said to be "five feet tall" or "six feet tall" is only incompatible because John can only be a single number of feet tall. If the attribute were a possession as in "he has a dog" or "he also has a cat", a model inconsistency would not happen. In other words, the issue of model inconsistency has to do with our model of the domain element (John) and not with the ascription of different range elements (measurements such as "five feet tall" or "six feet tall").
LipNet
LipNet is a deep neural network for audio-visual speech recognition (ASVR). It was created by University of Oxford researchers Yannis Assael, Brendan Shillingford, Shimon Whiteson, and Nando de Freitas. The researchers stated that could match mouth movements to text with 93 percent accuracy, though it was criticized for its test using a limited dataset of words and grammar. It was used in Nvidia's autonomous "backseat driver" prototype Co-Pilot.
Zvi Mowshowitz
Zvi Mowshowitz is an American writer and member of the rationalist community who primarily discusses new developments in artificial intelligence. He is a former competitive Magic: The Gathering player and was CEO of MetaMed. == Career == Mowshowitz is an alumnus of Columbia University and holds a bachelor's degree in mathematics. He co-founded and was the CEO of MetaMed, a medical research analysis firm. He has worked at Jane Street Capital, and has worked for the gambling industry in Las Vegas. He attempted to launch a blockchain game, Emergents, in 2020. === Magic: The Gathering === Mowshowitz held a developer intern position at Wizards of the Coast R&D in 2005. He created the deck TurboZvi. His first-place finishes at major competitions were the 1999 World Championships as part of the four-person United States national team, the 2001 Pro Tour Tokyo, and two 2003 Grand Prix. He has placed in the top eight of four Pro Tours, and earned over $140,000 playing Magic competitively. In 2007, Mowshowitz was elected into the Magic Hall of Fame. Last updated: 12 May 2013Source: Wizards.com Mowshowitz has written about Magic for several outlets, including the official Magic website. === Later career === Mowshowitz is on the board of directors for the Center for Applied Rationality, and is a member of the rationalist community. He also founded Balsa Research, a nonprofit think tank which advocated for the repeal of the Jones Act, increasing the housing supply, and reform of the National Environmental Policy Act. In 2023, Mowshowitz wrote an article for Vox on the topic of artificial intelligence safety. Mowshowitz has a blog on Substack under the name "Don't Worry about the Vase". He has written on topics such as artificial intelligence, economics, and the COVID-19 pandemic. == Personal life == Mowshowitz is the son of American biochemist Deborah Mowshowitz. His parents have both worked as Columbia University professors.
Integrated writing environment
An integrated writing environment (IWE) is software that provides comprehensive writing and knowledge management functionality for writers and information workers. IWEs enable writers and information workers to perform a variety of tasks related to the document in the IWE in a single environment. This provides a distraction-free workspace and streamlined writing experience. IWEs provide similar efficiency and functionality benefits to writers and information professionals that integrated development environments (IDEs) provide to software developers. == Overview == IWEs are designed to maximize productivity and help improve the quality of written work by integrating together tools that allow users to work effectively in a single application. The IWE features may include integrated content search, reversion management, outlining, note management, and reference management, as may be suitable for the target field of use. == List of IWEs == Celtx This IWE is intended for screenplay writers and has screenplay writing and management tools. Celtex provides tools for the pre-production work phase, story development, storyboarding, script breakdowns, production scheduling, and reports. Scrivener This IWE targets novel, research paper, and script writing. Scrivener provides tools to organize notes and research documents for easy access and referencing. After completing the writing, Scrivener allows the user to export the document to formats supported by common word processors, such as Microsoft Word. TeXstudio This IWE targets LaTeX documents and provides interactive spelling checker, code folding, and syntax highlighting.
Business rules engine
A business rules engine is a software system that executes one or more business rules in a runtime production environment. The rules might come from legal regulation ("An employee can be fired for any reason or no reason but not for an illegal reason"), company policy ("All customers that spend more than $100 at one time will receive a 10% discount"), or other sources. A business rule system enables these company policies and other operational decisions to be defined, tested, executed and maintained separately from application code. Rule engines typically support rules, facts, priority (score), mutual exclusion, preconditions, and other functions. Rule engine software is commonly provided as a component of a business rule management system which, among other functions, provides the ability to: register, define, classify, and manage all the rules, verify consistency of rules definitions (”Gold-level customers are eligible for free shipping when order quantity > 10” and “maximum order quantity for Silver-level customers = 15” ), define the relationships between different rules, and relate some of these rules to IT applications that are affected or need to enforce one or more of the rules. == IT use case == In any IT application, business rules can change more frequently than other parts of the application code. Rules engines or inference engines serve as pluggable software components which execute business rules that a business rules approach has externalized or separated from application code. This externalization or separation allows business users to modify the rules without the need for IT intervention. The system as a whole becomes more easily adaptable with such external business rules, but this does not preclude the usual requirements of QA and other testing. == History == An article in Computerworld traces rules engines to the early 1990s and to products from the likes of Pegasystems, Fair Isaac Corp, ILOG and eMerge from Sapiens. == Design strategies == Many organizations' rules efforts combine aspects of what is generally considered workflow design with traditional rule design. This failure to separate the two approaches can lead to problems with the ability to re-use and control both business rules and workflows. Design approaches that avoid this quandary separate the role of business rules and workflows as follows: Business rules produce knowledge; Workflows perform business work. Concretely, that means that a business rule may do things like detect that a business situation has occurred and raise a business event (typically carried via a messaging infrastructure) or create higher level business knowledge (e.g., evaluating the series of organizational, product, and regulatory-based rules concerning whether or not a loan meets underwriting criteria). On the other hand, a workflow would respond to an event that indicated something such as the overloading of a routing point by initiating a series of activities. This separation is important because the same business judgment (mortgage meets underwriting criteria) or business event (router is overloaded) can be reacted to by many different workflows. Embedding the work done in response to rule-driven knowledge creation into the rule itself greatly reduces the ability of business rules to be reused across an organization because it makes them work-flow specific. To create an architecture that employs a business rules engine it is essential to establish the integration between a BPM (Business Process Management) and a BRM (Business Rules Management) platform that is based upon processes responding to events or examining business judgments that are defined by business rules. There are some products in the marketplace that provide this integration natively. In other situations this type of abstraction and integration will have to be developed within a particular project or organization. Most Java-based rules engines provide a technical call-level interface, based on the JSR-94 application programming interface (API) standard, in order to allow for integration with different applications, and many rule engines allow for service-oriented integrations through Web-based standards such as WSDL and SOAP. Most rule engines provide the ability to develop a data abstraction that represents the business entities and relationships that rules should be written against. This business entity model can typically be populated from a variety of sources including XML, POJOs, flat files, etc. There is no standard language for writing the rules themselves. Many engines use a Java-like syntax, while some allow the definition of custom business-friendly languages. Most rules engines function as a callable library. However, it is becoming more popular for them to run as a generic process akin to the way that RDBMSs behave. Most engines treat rules as a configuration to be loaded into their process instance, although some are actually code generators for the whole rule execution instance and others allow the user to choose. == Types of rule engines == There are a number of different types of rule engines. These types (generally) differ in how Rules are scheduled for execution. Most rules engines used by businesses are forward chaining, which can be further divided into two classes: The first class processes so-called production/inference rules. These types of rules are used to represent behaviors of the type IF condition THEN action. For example, such a rule could answer the question: "Should this customer be allowed a mortgage?" by executing rules of the form "IF some-condition THEN allow-customer-a-mortgage". The other type of rule engine processes so-called reaction/Event condition action rules. The reactive rule engines detect and react to incoming events and process event patterns. For example, a reactive rule engine could be used to alert a manager when certain items are out of stock. The biggest difference between these types is that production rule engines execute when a user or application invokes them, usually in a stateless manner. A reactive rule engine reacts automatically when events occur, usually in a stateful manner. Many (and indeed most) popular commercial rule engines have both production and reaction rule capabilities, although they might emphasize one class over another. For example, most business rules engines are primarily production rules engines, whereas complex event processing rules engines emphasize reaction rules. In addition, some rules engines support backward chaining. In this case a rules engine seeks to resolve the facts to fit a particular goal. It is often referred to as being goal driven because it tries to determine if something exists based on existing information. Another kind of rule engine automatically switches between back- and forward-chaining several times during a reasoning run, e.g. the Internet Business Logic system, which can be found by searching the web. A fourth class of rules engine might be called a deterministic engine. These rules engines may forgo both forward chaining and backward chaining, and instead utilize domain-specific language approaches to better describe policy. This approach is often easier to implement and maintain, and provides performance advantages over forward or backward chaining systems. There are some circumstance where fuzzy logic based inference may be more appropriate, where heuristics are used in rule processing, rather than Boolean rules. Examples might include customer classification, missing data inference, customer value calculations, etc. The DARL language and the associated inference engine and editors is an example of this approach. == Rules engines for access control / authorization == One common use case for rules engines is standardized access control to applications. OASIS defines a rules engine architecture and standard dedicated to access control called XACML (eXtensible Access Control Markup Language). One key difference between a XACML rule engine and a business rule engine is the fact that a XACML rule engine is stateless and cannot change the state of any data. The XACML rule engine, called a Policy Decision Point (PDP), expects a binary Yes/No question e.g. "Can Alice view document D?" and returns a decision e.g. Permit / deny.