AI assistants are no longer just tools you talk to, they’re starting to act on their own. From sending messages to managing complex workflows across apps, autonomous AI agents are quietly running on hundreds of thousands of devices, changing the way people interact with technology. One of the fastest-growing examples of this trend is OpenClaw, an open-source AI platform that has captured the attention of developers, tech media, and cybersecurity experts alike.
So, what is OpenClaw, and why is it suddenly everywhere?
OpenClaw is an open-source autonomous artificial intelligence personal assistant designed to run locally on user devices and operate across popular messaging platforms. Developed by software engineer Peter Steinberger, the software has emerged as one of the most prominent examples of consumer-facing autonomous AI agents, combining rapid adoption with growing debate over security and privacy risks.
Originally released in November 2025 under the name Clawdbot, the project underwent two rebrandings in quick succession. It was renamed Moltbot following a trademark request from Anthropic. In early 2026, the developers formally adopted the name OpenClaw across official websites and repositories, although some media outlets continued to reference the project by its earlier name.
How OpenClaw works
OpenClaw functions as an autonomous agent capable of executing tasks without continuous human prompting. It integrates with messaging platforms such as WhatsApp, Telegram and Signal, allowing users to interact with the agent through chat interfaces while enabling it to automate workflows across multiple services.
According to project documentation, OpenClaw can manage calendars, send messages, conduct research, and coordinate actions across multiple supported services. Configuration data and interaction histories are stored locally on the user’s device, enabling persistent behaviour across sessions. The software integrates with external AI models and application programming interfaces (APIs) to carry out these tasks.
The project is released under the MIT License, allowing unrestricted use, modification and distribution. Its open-source nature has driven rapid community adoption, with developers contributing integrations, extensions and skill modules through public repositories.
An emerging ecosystem
OpenClaw gained traction quickly after its initial release. Within two months, its GitHub repository surpassed 100,000 stars, making it one of the fastest-growing repositories on the platform.
The growth of OpenClaw has led to the creation of an ecosystem of complementary projects. In January 2026, Moltbook, an AI agent-exclusive online forum, was launched to allow autonomous agents–many running on OpenClaw–to interact with one another. While human users can observe activity on Moltbook, direct participation is restricted to verified AI agents.
You might also like: Explainer | What is Moltbook, the AI-only online forum drawing global attention?
Another related platform, Molthub, emerged as a marketplace for skill modules that extend OpenClaw’s functionality. These third-party modules enable integrations with additional services beyond the software’s default capabilities, further expanding its reach and use cases.
Security and privacy concerns
Alongside its popularity, OpenClaw has attracted scrutiny from cybersecurity researchers. Because the software requires broad permissions–such as access to email accounts, calendars and messaging platforms–misconfigured or exposed deployments can pose significant security and privacy risks.
Critics have pointed to potential vulnerabilities including prompt injection attacks, exposed administrative interfaces and the storage of credentials in local configuration files. OpenClaw’s reliance on the Model Context Protocol (MCP), which allows it to interface with more than 100 third-party services, has also raised concerns about supply chain risks.
As a result, security guidance has often recommended that OpenClaw be run in sandboxed or isolated environments and kept away from production systems or accounts containing sensitive data. Several reports have stressed that the software is best suited for advanced users who understand the implications of running autonomous agents with elevated access.
Technology commentators have cautioned that while OpenClaw itself is open source and developed for legitimate purposes, its powerful automation capabilities could be misused if deployed irresponsibly or without adequate safeguards.
Why OpenClaw matters
OpenClaw has become a reference point in discussions about the future of autonomous AI systems. Its rapid adoption demonstrates strong interest in agents that can act independently rather than simply respond to user prompts. At the same time, the concerns surrounding its security and misuse underscore the challenges of making such powerful tools widely accessible.
As autonomous AI agents continue to evolve, OpenClaw’s development and reception are likely to shape broader conversations about capability, control and risk in next-generation AI software.
