Is Claude stirring up Silicon Valley AI to start taking a share of the enterprise software market?
Is it disrupting traditional enterprise solutions and challenging established players?
With its advanced capabilities, AI is beginning to reshape how businesses operate, and Claude is at the forefront of this wave.
Will this lead to a major shift in the industry, or will it face resistance from existing giants?
The competition is heating up as AI continues to evolve and penetrate deeper into enterprise applications.

In the wave of rapid advancements in artificial intelligence (AI) technology, a fierce battle over “who will dominate the next generation of enterprise software” is quietly unfolding.

Recently, the AI assistant Claude launched by American AI startup Anthropic has caused a market upheaval with its powerful programming capabilities, industry plugin ecosystem, and deep integration into enterprise workflows. This has not only led to consecutive declines in tech stocks but also sparked widespread industry discussion about the future of the SaaS (Software as a Service) business model.

User Enthusiasm, Market Anxiety

Recently, Anthropic released its flagship AI model Claude Opus 4.6, surpassing previous versions in coding ability. Moreover, Opus 4.6 can apply its enhanced features to a range of daily tasks: running financial analyses, conducting research, and creating and editing documents, spreadsheets, and presentations. In Cowork environments, Claude can autonomously perform multiple tasks. In Claude Code, users can also assemble intelligent agent teams to collaborate on tasks.

Two days later, Anthropic launched Claude Opus 4.6 “Fast Mode,” which increased speed by 2.5 times.

Boris Cherny, head of Claude Code, stated that the team has been using this tool for development over the past few weeks, “which is undoubtedly a huge breakthrough for me personally.”

Anthropic released a preview of Claude Code in February 2025, officially opened Claude Code to the public in May; in September, Claude Sonnet 4.5 was released, and Claude Code 2.0 was launched, no longer limited to command line interface, allowing users to save states at any time while AI automatically modifies code. The underlying framework for building Claude Code was also officially opened to developers for customization. In January 2026, Anthropic further lowered the entry barrier by including Claude Code access in the Team plan.

This series of initiatives rapidly expanded the user base. Claude Code not only became popular among programmers but also attracted many non-technical users. On social media, there are stories of people who have never learned programming sharing their experiences of successfully developing their first application with Claude Code, using it for health data analysis, expense reporting, and other tasks.

Based on this trend, Anthropic quickly incubated a derivative product, Claude Cowork. Cherny revealed that, leveraging Claude Code’s capabilities, the entire Cowork project from conception to launch took only about 10 days.

Repeated capability upgrades of Anthropic’s models have caused a decline in the stock prices of U.S. software companies. Market anxiety has spread quickly—if enterprises can use AI to autonomously build customized tools, is there still a need to purchase standardized SaaS products?

Market research firm Analytics Insight pointed out that more and more developers are embedding models like Claude directly into their products, which could weaken the existing advantages and user stickiness of traditional SaaS vendors in data analysis and research workflows.

Thomas Shipp, head of stock research at LPL Financial, said, “People will think, since AI can significantly reduce the time needed to develop these systems internally, why should I pay for ready-made software? Moreover, with the release of Cowork and other products—applications that can access file reading and editing permissions—tech-savvy users now have the ability to replace existing workflows.”

Jensen Huang Supports “AI + Software”

In fact, AI’s involvement in the software field has already begun. In 2021, OpenAI launched Codex, which demonstrated the ability to generate executable code from natural language, leading to a series of programming assistance tools. But at that time, AI mainly played a supporting role, helping developers complete repetitive coding tasks faster, rather than restructuring entire business processes.

On the same day that Anthropic released Claude Opus 4.6, OpenAI also officially launched GPT-5.3-Codex. Codex can operate automatically without prompts, handling tasks such as problem routing, alert monitoring, and CI/CD. With its built-in work tree and cloud environment, intelligent agents can work in parallel across multiple projects, reducing development cycles from weeks to days. OpenAI also introduced a dedicated Codex App equipped with multi-agent command centers and localized integrations.

Today’s AI tools demonstrate more systematic capabilities.

In response to market anxiety, NVIDIA CEO Jensen Huang has publicly expressed different views multiple times. On February 4, at an industry forum, he stated, “Some believe that software tools are declining and will be replaced by AI… this is the most illogical thing in the world, and time will prove it.”

He further explained: “Software is a tool, and AI will use these tools, not reinvent them. We are迎接 the greatest opportunity in software history. For the first time, software is no longer just a tool. For example, Excel is a tool; now, software begins to use tools—these AIs will use Excel. Therefore, I believe this new era of software contains incredible opportunities.”

Analysts from industry research firm Aurelion Research also said that the recent sell-off was “emotion-driven,” and as enterprises gradually see measurable returns from AI, this sentiment may “return to normal.”

Nick Dempsey, head of media equity research at Barclays, pointed out that whether general AI models can truly become viable substitutes with industry expertise remains doubtful.

Li Bojie, co-founder and chief scientist of Pine AI, told The Paper that Claude’s recent release reflects that AI code production capabilities are becoming increasingly strong. However, this does not mean AI agents can directly replace SaaS, but reveals a development trend: as AI capabilities continue to upgrade, the market space for traditional SaaS industries will inevitably be compressed.

“In fact, AI front-line workers have long been aware of this phenomenon, but market reactions have been relatively lagging,” Li said. “This means that only those software companies that actively leverage AI to enhance their capabilities and fully utilize data advantages will be able to survive better in the future.”

Where is the Future of Software?

So, will intelligent agents truly overturn the underlying logic of the software industry?

Tan Jian, associate professor at Beijing University of Posts and Telecommunications’ School of Digital Media and Design Arts, believes that rather than challenging the traditional SaaS product logic, intelligent agents represent a “value return.” In his view, they are pulling SaaS back from “function tools” to “service commitments,” rewriting human-computer interaction and pricing models.

Tan Jian pointed out that the core of Claude Coworker is not to provide more features but to deliver “ready-to-use results.” Its plugins essentially package job SOPs (Standard Operating Procedures), tool connections, and trigger commands into reusable capabilities, which is not fundamentally different from the goal of traditional SaaS to pursue standardized output: “In the past, employees clicked buttons and learned systems; now, users define goals, and the system completes them.”

In Tan Jian’s view, in the short term, “AI cannot replace” the core markets of traditional SaaS. The key is “trust and responsibility chain.” If an intelligent agent makes a mistake, the impact is often greater than that of a human, but accountability is difficult to quickly assign. If such cases occur, it will significantly raise enterprises’ demands for “auditability, rollback, and accountability.”

Looking ahead, Tan Jian believes that as intelligent agents become more widespread, software pricing may shift from per-user charges to results-based payments. The industry will also differentiate into “tool-based SaaS replaced at the front end” and “value-result platform” service SaaS, with the latter being the only way to truly stay on the table in the AI era.

Li Bojie sees future competitive barriers in the software industry as becoming more segmented: on the ToB (enterprise) side, the focus is on data accumulation and domain knowledge; on the ToC (consumer) side, it will still need to return to traditional internet strategies, where product capabilities and operational skills will deeply influence competitiveness.

The rise of Claude is less an “impact” on software and more an impetus for the industry to upgrade. AI does not negate the value of software but redefines “how software should be used.” As Jensen Huang said, we are entering a new era of “AI-using software,” where humans set goals, AI dispatches tools, and traditional software becomes the “infrastructure” called upon by AI.

In this transformation, no one is destined to be eliminated—only those who fail to evolve in time. For enterprises, the key may not be whether they adopt Claude or Codex, but whether they can leverage AI to unleash their own value. For the software industry, the real challenge has just begun: how to continue being the indispensable “tool provider” in the age of intelligence.

(Source: The Paper)