Goldman Sachs's judgment: Claude Cowork becomes the first "trusted use case" for agent workflow, OpenClaw demonstrates the future direction of human-machine interaction.

AI applications are entering a critical next stage - comprehensive penetration from developer tools to knowledge workers. Goldman Sachs' latest research report outlines a clear path for the scale landing of general intelligence tools through two landmark products - Claude Cowork and OpenClaw. According to a report on March 7th, Goldman Sachs pointed out in its latest research report that AI as the next stage of general intelligence layer applications has emerged. Anthropic's Claude Cowork and open-source agent OpenClaw are the two core data points defining this trend. The report stated that Claude Cowork is the first "trusted case for agent workflow for knowledge workers," which is reshaping the value distribution in the SaaS industry. It will not directly replace existing record systems (such as CRM or ERP), but will push market share towards SaaS providers that can effectively integrate AI capabilities. Meanwhile, the viral spread of OpenClaw not only demonstrates a new form of future human-machine interaction, but its unique hosting economics is breaking the monopoly of hyperscalers, directing massive new inference compute demand directly to alternative edge computing and VPS providers like DigitalOcean and Cloudflare. Claude Cowork: The first trusted case of agent workflow for knowledge workers On January 12th, Anthropic officially launched Claude Cowork - an agent tool embedded in the Claude desktop application, designed for non-developers, capable of automating file management and task processes covering multiple steps such as file organization, document processing, and workflow management. On January 30th, Anthropic further released a series of Cowork plugins targeting "business functions," covering scenarios such as sales and marketing, finance, legal, customer support, data analysis, productivity and enterprise search, and product management. Goldman Sachs believes that Cowork's core value lies in filling the critical gap between developer-specific agent tools and the daily use of chat Siasun Robot & Automation by ordinary knowledge workers. Software engineers have seen productivity gains through terminal agent loops for over a year, while knowledge workers' AI interactions have long been limited to chat interfaces. Cowork extends this ability to non-technical personnel through an application interface integrated with Excel and standard business tools. Specific landing scenarios include: accounting personnel categorizing credit card transactions, HR teams compiling tax reports and tracking PTO, CFO accelerating project-specific DCF analysis. In the digital agent customer base, the marketing scenario is the most prominent - customers are integrating Klaviyo and automating the entire reporting process. Cowork's core competitive advantage lies in its position as an intelligent layer across SaaS stacks: it is not tied to any single SaaS product, can switch between tools freely, and infer data across multiple systems' records and the "glue layer." Goldman Sachs' industry research shows that Anthropic is in a clearly leading position in tool capabilities and underlying model intelligence. Cowork will not replace system records, but will reshape SaaS value distribution Goldman Sachs explicitly stated that tools like Cowork will not replace core system records such as CRM, ERP, and companies do not have the intention to dismantle existing SaaS stacks. However, as AI becomes a key differentiating factor, wallet share will concentrate on SaaS vendors that can provide the most effective AI workflows. In practice, customers are replacing lightweight "glue" software with Cowork - tools that primarily move data between systems rather than core platforms. Conversations with Goldman Sachs partners show that companies still prefer to pay third-party software vendors for security, governance, and operational control. A case study of cooperation between Intuit and Anthropic is a prime example of this logic. On February 24th, Intuit announced a partnership with Anthropic: Intuit will gain access to Anthropic models and custom AI agents through the Claude Agent SDK, while Anthropic will promote Intuit products through MCP integration on the Cowork, Claude for Enterprise, and Claude.ai platforms. Goldman Sachs believes that this partnership logic is similar to Intuit's previous $100 million annual agreement with OpenAI: base models do not need to build custom workflows, but monetize through model capabilities and reach; Intuit retains control of the data and maintains its unit economics. Goldman Sachs notes that the potential risk lies in adding another layer between customers and Intuit, which may lead to long-term abstraction of value; but the potential reward is better customer acquisition, similar to the partnership model with search engines in the Web 2.0 era. It is worth noting that on March 5th, OpenAI introduced GPT-5.4 - its first universal model with native computer usage capabilities, similar to Claude Cowork, which can operate computers and execute agent workflows across applications. OpenAI particularly emphasizes that the model consumes about 47% fewer tokens in some tasks compared to its predecessor models, while maintaining the same accuracy. OpenClaw: Breaking contextual constraints, showing new directions for future human-machine interaction and compute demand Goldman Sachs believes that OpenClaw, by giving AI agents complete computer control and infinite memory, achieves viral growth in consumer-side applications; its unique local and low-cost VPS hosting model is becoming a significant advantage for edge computing providers. OpenClaw (formerly Moltbot and Clawdbot) was first released in November 2025 and achieved "viral" spread in late January. It is an open-source, self-hosted, AI agent that runs 24/7 on the user's local computer, able to receive commands via personal messaging platforms like iMessage, WhatsApp, Google Chat, and execute any task that a human can, including opening apps, browsing the web, and completing transactions. OpenClaw addresses a key technical bottleneck that limits most AI agents - the context window problem. Its solution is based on a file-based memory system: all agent activities are stored in Markdown files on the local computer, and the agent loads the memory for that day and the context from the previous day at the start of each session. Users can also establish long-term memories through commands (such as "always prioritize OpenTable when booking a restaurant"), forming a persistent preference database. Matthew Prince, co-founder and CEO of Cloudflare, pointed out that the potential trajectory of OpenClaw over the next 3 years may mirror the path of ChatGPT over the past 3 years: currently catalyzing a large number of experiments, but, similar to the early days of ChatGPT, it may take several years to achieve sustained scalability for daily use. OpenClaw's compute economics: structural opportunities for edge computing and VPS providers OpenClaw's hosting economics have important implications for investors. Due to the cost volatility of hyperscale cloud service providers, most users choose to run OpenClaw on home hardware (Mac Mini, old laptops) or low-cost VPS providers like Cloudflare and DigitalOcean. Some users have incurred thousands of dollars in token consumption in the first few days of use. Many users are adopting a layered model architecture: where advanced models orchestrate, smaller models execute sub-tasks, lowering the cost per query. DigitalOcean's data is the most intuitive: in the days following the release of OpenClaw, customers deployed nearly 30,000 native one-click OpenClaw Droplets on DigitalOcean, with thousands more activations. In DigitalOcean's AI customer revenue, inference services continue to increase - with AI customer revenue ARR reaching $120 million in December 2025, where non-bare metal business accounts for 70%, and inference services growing at a YoY rate of 254%. Cloudflare has not yet seen significant impacts on logo growth or network traffic from OpenClaw, but Goldman Sachs believes that this trend further confirms Cloudflare's leading cost-effective advantages in edge computing and Workers software. This article is from Wall Street SeeNews, GMTEight editor: Yan Li.