Key points:

• X has introduced a new video recording and editing tool for iOS users.
• The editor includes multilingual captions and a green-screen feature.
• The company aims to encourage creators to publish original content on X.
• More improvements to the video editor are expected in the coming weeks.
• X continues to battle bots, recycled videos, and competition from TikTok, YouTube, and Meta.
• Video posts now account for nearly half of all impressions on X.

BERLIN – X has introduced a new video recording and editing tool for its iOS application as the social media platform seeks to encourage creators to publish original content instead of reposting videos from other platforms.

According to TechCrunch, the new editor is currently available only on iOS. The Android version has not yet received the feature because the application is still being rebuilt.

The latest update adds several editing tools designed to simplify video production directly within X.

What are the tools of the Editor?

The editor allows creators to add captions in multiple languages while also giving them the ability to customise the appearance of those captions.

Another notable addition is a green-screen feature that enables users to place photos stored on their phones or existing X posts behind their videos, offering greater flexibility for content creation.

X’s Head of Product, Nikita Bier, said supporting original creators and providing better production tools remain among the company’s top priorities. He added that additional improvements to the editor will be introduced over the coming weeks.

What does X want from this editor?

X Wants More Original Content

The new editor forms part of X’s broader strategy to increase the amount of original content published on the platform.

Bier has previously said that many of the platform’s largest accounts frequently repost recycled or stolen videos that originally gained popularity on other social media platforms several years ago.

While improved editing tools may encourage some creators to produce exclusive content for X, reposting existing videos continues to be an easier method for many accounts to attract views and generate revenue.

The growing competition in the market

X continues to face intense competition from TikTok, YouTube and Meta, all of which offer established creator programmes and more consistent monetisation opportunities.

Industry observers believe creators are unlikely to publish content exclusively on X unless the platform can provide dependable earnings alongside a large and active audience.

The company also trails some competitors in copyright protection. Meta allows eligible Reels creators to identify copied videos and either block them or add attribution links, while YouTube provides systems for detecting and requesting the removal of unauthorised uploads.

How bots cause a nuisance?

Automated accounts remain another obstacle to X’s efforts to build a stronger creator ecosystem.

Bier said in April that the platform was detecting and suspending around 208 bots every minute. He also revealed that nearly half of X’s product team had been focused on reducing spam across the platform.

The challenge is not unique to X. Reddit has introduced AI-powered tools to tackle increasing spam generated by large language models, while Digg recently shut down its application after saying it lacked the resources needed to combat widespread spam.

The role of videos on X

According to Bier, video posts now account for nearly half of all impressions on X, highlighting the format’s growing importance.

The company has recently introduced several creator-focused video features, including React with Video, which enables users to create green-screen or split-screen responses to existing posts.

Analysts say X will likely need to combine improved creation tools with stronger copyright protection, consistent monetisation and continued action against bots if it hopes to attract more creators producing exclusive content for the platform.

Key points: 

• Multi-Device Support: Anthropic is launching the Claude Cowork expansion, bringing the AI agent to iOS, Android, and web browsers. 

• Cloud-Powered Sessions: Therefore, tasks now run autonomously in the cloud, meaning operations continue even when laptops are closed. 

• Non-Coding Focus: Newly released data reveals that over 90% of tasks are used for general business operations rather than programming. 

• Rolling Beta: Consequently, the update is rolling out in beta over the coming weeks, starting with Claude Max subscribers. 

ISLAMABAD – The official Claude Cowork expansion has arrived to transform how professionals manage automated workflows across multiple devices. This update shifts the multi-step AI assistant from a desktop-only application into a flexible, cloud-based cross-platform tool.

As a result, users can now assign complex tasks, track active progress, and approve finished files from any web browser or mobile phone. The feature originally launched as a standalone desktop app. However, beta access is now rolling out directly inside the standard Claude apps for iOS and Android.

Seamless Background Monitoring and Cloud Control 

Because of this cloud migration, the Claude Cowork expansion completely untethers the AI from your local hardware. Sessions now run securely on Anthropic’s remote servers. For this reason, automated tasks will continue executing even if a user closes their laptop or goes entirely offline. For example, you can easily initiate a research project at your desk, leave the office, and monitor live progress directly from your smartphone. 

Furthermore, if the AI agent hits a bottleneck or requires specific human guidance, it sends an interactive notification directly to the user’s phone. The user can then instantly redirect the workflow or input required data. Nevertheless, Anthropic emphasizes that the human remains entirely in control, as no emails are sent or files finalized without explicit user approval. 

“The desktop application will still serve as the main hub for deep local work, as it retains direct access to local files,” Anthropic clarified. “However, the web and mobile expansion opens the tool to users who cannot install desktop apps.” 

AI Agents Pivot Toward General Administration 

This widespread rollout highlights a major shift in how tech companies position AI agents. While tools like OpenAI’s Codex started primarily as software development assistants, Anthropic is steering Cowork to be a general administrative partner. In addition, Anthropic released an aggregated study of 1.2 million Cowork sessions across 600,000 organizations to prove this trend. 

The data confirms that everyday office tasks dominate the platform. For instance, business operations accounted for 33.4% of usage, including compiling spreadsheets and formatting HR checklists. Meanwhile, content creation represented 16.4% of tasks. Finally, software development accounted for only 8.7% of activity. Therefore, the findings show that more than 90% of overall usage occurs outside of traditional programming. 

Project Technical Overview 

Frequently Asked Questions (FAQs) 

What does the Claude Cowork expansion allow users to do? 

It allows professionals to delegate multi-step administrative tasks to Claude and manage them across mobile devices and web browsers without needing the desktop app open or online. 

Can Claude Cowork run while my computer is turned off? 

Yes, because of the new cloud infrastructure, scheduled or active tasks run completely in the background even if your laptop is closed, turned off, or disconnected from the internet. 

Do I need to know how to code to use Claude Cowork? 

Not at all. While it shares an architecture with coding tools, company data shows over 90% of its use cases are for non-technical office work, like managing spreadsheets, organizing files, and drafting documents.

Read more: Google fixes Dialogflow CX vulnerability saving exposed AI chatbot data. 

Key points:

• Google has fixed a critical security vulnerability affecting its Dialogflow CX AI platform.
• Researchers said the flaw could have allowed attackers to steal chatbot conversations and launch phishing attacks.
• The vulnerability required the dialogflow.playbooks.update permission to be exploited.
• Google said the issue has been fully mitigated and found no evidence of customer compromise.
• Security experts have advised organisations to review Playbook updates, Code Blocks and system logs as a precaution.

KARACHI – Google has patched a critical security vulnerability in its Dialogflow CX artificial intelligence platform that researchers say could have allowed attackers to steal sensitive chatbot conversations and conduct phishing campaigns.

The flaw, dubbed “Rogue Agent” by cybersecurity firm Varonis, affected Dialogflow CX, Google’s enterprise platform used to build AI-powered customer service agents, healthcare assistants, financial services chatbots and other conversational AI applications.

How bad was the vulnerability?

According to Varonis Threat Labs, the vulnerability stemmed from a permission boundary issue that could have enabled attackers to inject persistent malicious code into AI agents through the platform’s Code Blocks feature.

Researchers said exploitation required the dialogflow.playbooks.update permission on a Dialogflow agent. Once obtained, an attacker could potentially modify the execution pipeline, allowing malicious code to intercept conversations, extract sensitive information and insert fraudulent prompts aimed at stealing user credentials.

What Google has to say about the vulnerability?

Google confirmed that the vulnerability has been fully addressed and said customers do not need to take any action.

The issue was privately disclosed to Google by Varonis in November 2025. Google released an initial patch in April 2026 before implementing a complete fix last month, resolving all affected components.

What is Dialogflow CX AI?

CX allows developers to add custom Python-based Code Blocks that process user input, interact with APIs and automate workflows. According to the researchers, these Code Blocks execute within Google’s Cloud Run environment, which allows outbound internet connectivity by default.

Varonis said that if an attacker gained permission to modify a Playbook’s Code Block such as through a compromised employee account they could replace part of the execution process with malicious code capable of accessing conversation history and session data or presenting fake authentication requests to users.

The researchers noted that although initial access would still be required, the necessary permission could be granted at the project level, meaning the attack was not limited to top-level administrators.

Were customers affected?

While Google said there is no evidence that customers were affected, Varonis recommended that organisations review Playbook update logs, inspect Code Blocks for unauthorised changes and investigate failed user requests that could indicate malicious activity.

What should organisations do to address this?

Security researchers also urged organisations to treat AI infrastructure as part of their overall cloud security strategy, warning that configuration weaknesses in AI services can create opportunities for attackers even when core platforms are secure.