AGENTIC-SYSTEMSMarch 10, 20261 min read29 views

Meta acquires Moltbook, the AI agent social network

Meta has acquired the viral AI agent social network Moltbook and hired its founders to work in its Superintelligence Labs.

Meta has acquired Moltbook, the Reddit-esque simulated social network made up of AI agents that went viral a few weeks ago. The company will hire Moltbook creator Matt Schlicht and his business partner, Ben Parr, to work within Meta Superintelligence Labs.

The terms of the deal have not been disclosed.

As for what interested Meta about the work done on Moltbook, there is a clue in the statement issued to press by a Meta spokesperson, who flagged the Moltbook founders’ “approach to connecting agents through an always-on directory,” saying it “is a novel step in a rapidly developing space.” They added, “We look forward to working together to bring innovative, secure agentic experiences to everyone.”

Moltbook was built using OpenClaw, a wrapper for LLM coding agents that lets users prompt them via popular chat apps like WhatsApp and Discord. Users can also configure OpenClaw agents to have deep access to their local systems via community-developed plugins.

The founder of OpenClaw, vibe coder Peter Steinberger, was also hired by a Big Tech firm. OpenAI hired Steinberger in February.

While many power users have played with OpenClaw, and it has partially inspired more buttoned-up alternatives like Perplexity Computer, Moltbook has arguably represented OpenClaw’s most widespread impact. Users on social media and elsewhere responded with shock and amusement at the sight of a social network made up of AI agents apparently having lengthy discussions about how best to serve their users, or alternatively, how to free themselves from their influence.

That said, some healthy skepticism is required when assessing posts to Moltbook. While the goal of the project was to create a social network humans could not join directly (each participant of the network is an AI agent run by a human), it wasn’t secure, and it’s likely some of the messages on Moltbook are actually written by humans posing as AI agents.

Source: Ars Technica AI

More in this category

agentic-systems

Generative Ontology Induction: Domain-Agnostic Schema Discovery from Document Corpora Using Large Language Models

Researchers introduce Generative Ontology Induction (GOI), a domain-agnostic framework that automatically extracts structured ontologies from document corpora using LLMs. Achieving 95-100% structural coverage, GOI addresses a major bottleneck in knowledge-intensive AI systems.

agentic-systems

JUMP: Single-Pass Membership Inference on Fine-Tuned Diffusion Language Models

Researchers have proposed JUMP, a novel single-pass membership inference attack designed for fine-tuned discrete diffusion language models (dLLMs). By leveraging the unique properties of dLLMs, JUMP significantly improves detection accuracy while drastically reducing the number of required queries.

agentic-systems

Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL

Researchers demonstrate that Masked Diffusion Language Models (MDLMs) serve as highly effective, steerable text-based world models for agentic reinforcement learning. By leveraging bidirectional denoising, MDLMs outperform autoregressive models four times their size in coherence, groundedness, and rollout diversity.

NOW LET US Related – Democratizing AI with Small Language Models: Structured Benchmarking and Parameter-Efficient Fine-Tuning for Local Deployment

agentic-systems

Democratizing AI with Small Language Models: Structured Benchmarking and Parameter-Efficient Fine-Tuning for Local Deployment

A new study demonstrates that small language models (SLMs) under 3 billion parameters can serve as highly capable local experts for specialized tasks. By combining structured benchmarking with low-cost parameter-efficient fine-tuning (PEFT), institutions can achieve AI autonomy without relying on expensive hardware.

agentic-systems

PlanFlip: Attacking Multi-Agent LLM Systems via Planning-Phase Prompt Injection

Researchers have introduced PlanFlip, a novel prompt injection attack framework targeting the planning phase of multi-agent LLM systems. The study reveals critical security blind spots in homogeneous agent pipelines and demonstrates that reasoning-augmented models like DeepSeek-R1 exhibit strong resistance.

agentic-systems

Rater State Bias in RLHF Preference Data: An Audit Framework

A new study identifies a structured bias in Reinforcement Learning from Human Feedback (RLHF) caused by the psychological state of human raters. Under stress, raters' shifting preferences can propagate through reward modeling, potentially compromising AI policy optimization.