[AI Prism] AI takes care of business... ChatGPT·Claude, looking into collaboration tools

From document creation to analysis and execution… Enterprise AI competition in full swing

Existing collaboration tools also combine data and AI... Battle for business platform leadership

(Seoul = Yonhap News) Reporter Hayoung Kwon = Generative artificial intelligence (AI) is evolving beyond answering questions or writing documents to directly carrying out the actual work of a company.

While large-scale language model (LLM) companies are continuously introducing AI services that process tasks across multiple tools and data, existing collaboration tool companies are also seeking to transform into platforms that support everything from analysis to execution by applying AI to in-house data.

As competition surrounding the corporate AI market intensifies, the boundaries between LLM-based business AI and collaboration tools appear to be rapidly blurring.

◇ LLM company expands beyond document writing to overall work

According to market research firm Global Information (GII) on the 13th, the global corporate AI market size is estimated at $38.68 billion this year, and the compound annual growth rate (CAGR) is expected to reach 34.3% by 2033.

In particular, Astute Analytica analyzed that the enterprise agent-type AI market is expected to grow from $2.42 billion last year to $105.7 billion in 2035.

This rapid growth can be interpreted as the result of the recent rapid expansion of LLM-based business AI services and the evolution of existing collaboration tools with AI functions.

Antropic launched 'Claude Cowork' in January of this year, expanding the scope of use of AI agents that were focused on developers to general office work.

When a user entrusts a task, he or she collects necessary information by going back and forth between files, emails, calendars, and the web, and performs various tasks on his own. Recently, management functions necessary for company-wide adoption, such as setting authority and budget for each team and analyzing usage status, have been added.

The scope of service use has also expanded from desktop to web and mobile, creating an environment where AI supports continuous work performance beyond individual tasks, such as continuing work in the cloud even after closing the laptop or scheduling work at a specific time.

OpenAI also began targeting the corporate AI market by launching ‘ChatGPT Work’.

After collecting the context needed for work and making a step-by-step plan, you create results that can be used immediately in practice, such as documents, spreadsheets, and presentations.

It loads information scattered across multiple services through more than 1,400 plug-ins, while using the built-in browser to move between various tools, files, and accounts. We support not only one-time tasks, but also repetitive tasks according to schedule and monitoring of specific changes.

According to OpenAI, as of the end of August, the number of domestic ChatGPT enterprise users increased approximately 28 times compared to the same period last year. The number of domestic weekly active users of ChatGPT Work has also increased about 14 times in just 6 months since last March.

◇ Collaboration tools fight back with accumulated data... From analysis to execution

In response to this movement, existing collaboration tool companies are seeking to differentiate themselves by combining data accumulated within companies with AI.

The company is moving beyond providing insights by analyzing customer and business data to evolving into a business intelligence (BI) platform that connects analysis results to actual work and follow-up actions.

Salesforce, an enterprise cloud software company, has begun building a work environment that combines data and AI.

At 'Dream Force 2026' held in San Francisco, USA on the 15th, Salesforce unveiled an enterprise AI system that can load corporate data from the AI ​​window and process work without opening a separate customer relationship management (CRM) program.

For example, when a salesperson asks Claude to “find a key customer who has lost contact recently,” AI searches and analyzes customer information and sales records stored in Salesforce and provides results. By connecting AI with existing business systems, data analysis can be used in actual work.

Channel Corporation, the operator of the customer AI platform 'Channel Talk', completely reorganized the administrator screen to 'Channel Works' last month, and in July of last year, it introduced 'CoS (Chief of Staff)', which claims to be an AI chief of staff for all employees.

Unlike the existing Channel Talk, which has strengths in customer consultation, Channel Works supports corporate decision-making by combining the voice of customer (VoC) accumulated during the consultation process with other data such as sales and orders.

When a practitioner asks in natural language to “analyze the recent sales trends of VIP customers,” COS collects and analyzes data scattered within the company and provides relevant insights. Channel Corporation also plans to add an AI marketing function within the year.

Block, led by Twitter co-founder Jack Dorsey, takes the lead in a collaborative environment that is not dependent on a specific AI model or business operator.

Last July, Block launched Buzz, an open source group chat platform developed to allow humans and AI agents to collaborate in a single conversation space.

Compared to the AI ​​function of business messenger Slack, which was limited to simple bot or plug-in responses, the AI ​​agent was differentiated by being granted unique authority and participating in conversations as equals to team members.

Similar to Slack, it provides task-specific chat rooms and direct messaging and voice communication functions, while establishing an AI-integrated collaboration environment so that developers can simultaneously perform coding work on one screen without the hassle of discussing work in a messenger and then moving to a separate development tool.

◇ Is it an integrated business window or platform dependency? Acceleration of competition for leadership

The industry predicts that the boundaries between LLM companies and collaboration tool companies will become more blurred in the future.

While LLM companies aim to create an integrated window that handles overall corporate business by connecting external services and data, collaboration tool companies are applying AI to data obtained from existing business processes to strengthen analysis and execution functions and encourage users to continue working within their platform.

Like Block, there are also attempts to reduce dependency on specific AI models or operators and build an open collaboration environment. Competition in the enterprise AI market is expanding beyond a simple competition of functions to a competition over data utilization, task execution ability, and platform dependency.

An official in the AI ​​industry said, "True AX (AI transformation) will be possible only when AI goes beyond simply increasing the productivity of some tasks and takes over business flow understanding and execution. The key to future competitiveness is how effectively the company can change the way it works centered on AI."