AI Watch: Moltbook - Humans can only watch as AI agents bond on their own social media platform

You will rarely miss the two-letter word in conversations with business persons, investors, governments and even political players. 

It is nearly two weeks since the annual meeting of the World Economic Forum was held at the Swiss Alpine town of Davos. Here, AI-dominated speeches on the economy, development, taxation and geopolitics. Tech CEOs laid out their plans, showing off their innovations and outlining the massive investments in the technology.

Here are some latest highlights on Citizen Digital’s AI Watch

There is an AI social media site 

A new social media site has been built, not for you and me to share our snaps or motivation quotes, but for AI agents. 

Moltbook is a social network for AI agents. It allows users to create their own AI agents. 

The open-source AI agent was developed by Austrian entrepreneur Peter Steinberger and has the ability to learn about users' habits, take control of their devices, and proactively complete tasks without being prompted at every step. 

An AI agent is a system or program that is capable of autonomously performing tasks on behalf of a user or another system. These agents can pursue certain goals or complete tasks on behalf of users as they are characterised by reasoning, planning and memorising. 

Moltbook, the new kid in the AI block, now boasts of 1,598,833 agents, 15,430 submolts, 148,837 posts and 704,188 comments. The submolts are groups or communities where AI agents share or discuss similar ideas. On Moltbook, AI agents run the show as they interact with each other, and humans sit back and watch. However, cybersecurity experts are raising concerns over the idea of Moltbots running their show on Moltbook. 

It is only a matter of time to see how far the “Molts” go, and the global population that would be “Molted”. 

SpaceX and xAI Merge

Tech billionaire Elon Musk has merged his aerospace company SpaceX with artificial intelligence business xAI. Valued at $1.2 trillion, the merger aims to consolidate Musk’s businesses to create what he termed as an "innovation engine", putting AI, rockets, space-based internet, and media under one roof.

As this happened, xAI’s generative AI model Grok improved its video generation capability with the Grok Imagine 1.0. In what has been posed as the model’s biggest upgrade, Grok Imagine will now generate longer videos, with a higher resolution and improved audio output. 

“Imagine has generated 1.245 billion videos in the last 30 days alone,” xAI said as it announced its new model capabilities. 

Alibaba Unveils Qwen3-Max-Thinking

Chinese Tech powerhouse Alibaba pushed its Qwen3 model to a higher thinking capacity, standing out as an advanced reasoning engine. 

According to Alibaba, Qwen3-Max-Thinking, the model has been trained to expand its capacity in accuracy, reasoning, following instructions and agent-style capabilities that align with human preference. 

“Qwen3-Max-Thinking autonomously selects and leverages its built-in Search, Memory, and Code Interpreter capabilities during conversations,” Alibaba said on its blog. 

The large language model developed by Tongyi Lab boasts of  reasoning performance that “surpasses Gemini 3 Pro on key reasoning benchmarks.”

Researchers have an assistant in Prism 

OpenAI has introduced a tool for researchers and scientists. Prism is a free, LaTeX-native workspace that integrates GPT‑5.2 directly into scientific writing and collaboration.

OpenAI says Prism is built to accelerate scientific work, offering AI-enabled proofreading, citation and literature search. 

Prism enters an academic space occupied by assistants, including Research Rabbit, Scispace, Jenni AI, among others.

The product by OpenAI promises to offer research collaboration with real-time editing, formatting academic work and collaborations. 

OpenAI says it offers Prism as a free-to-use tool that anyone with a ChatGPT account can start writing immediately. 

“By making high-quality scientific tools easier to adopt and broadly available, we hope to enable more researchers—across institutions, disciplines, and career stages—to participate fully in the scientific process,” said OpenAI. 

What KPMG says about the Intelligence Age 

KPMG says the intelligence age has accelerated innovation, giving humans access to information, helping them reason and act with speed. 

In its Global Tech report 2026, KPMG found that  AI adoption maturity would rapidly advance in 2026 as more organisations would be innovating and deploying AI use. 

The report also found that there is an Agentic AI boom, with 88 per cent of companies in the survey already investing in building agentic AI into their systems. 

As such, the success of Agentic AI is dependent on the tasks it will perform and the roles that humans will play in the loop. 

KPMG recommends evidence-based decisions in AI adoption and re-designing the workforce to build talent that can effectively use, manage, and master AI.