Anthropic's chief business officer has drawn a clear line in the sand, stating the company will not engage in the AI industry's price competition and will instead double down on extracting maximum value for its enterprise clientele.

According to sources cited by Bloomberg on September 3rd, Anthropic is poised to secure a $15 billion credit facility ahead of its anticipated IPO. The financing round is spearheaded by Morgan Stanley, with Goldman Sachs, JPMorgan, and Citigroup slated for major roles. Barclays and Wells Fargo are also expected to participate.

The sheer scale of this financing underscores significant external confidence in the AI firm's public market prospects. In a recent interview, Anthropic's Chief Business Officer, Paul Smith, articulated his stance: "I have no interest in buying market share through price cuts. I'd rather focus on customers and how they extract maximum value from the models."

Analysts view this as a direct counter to OpenAI's strategy, which has involved multiple model price reductions over the summer.

Pricing Strategy: Value Over Volume, Declining to Follow Discounts

This summer, OpenAI cut prices across several of its AI models, a move widely interpreted as a response to competitive pricing pressures from developers in the US and China. As the current benchmark for capability, Anthropic has chosen a starkly different path.

Smith, who joined Anthropic over a year ago to lead enterprise sales, clarified that the core of their pricing strategy is customer value. He stated: "This isn't about price cuts or anything else; it's about how to create value for customers."

Just days after his interview, Anthropic released its new model, Fable 5.1, which emphasizes coding and scientific task capabilities while keeping pricing consistent with its predecessor.

However, the company has introduced targeted cost reductions where models recall previously processed information. While the input and output prices for Fable 5.1 remain unchanged, the cache read price has been slashed by 75%. According to Anthropic's calculations, this reduces the overall cost of typical workloads by roughly 25% compared to Fable 5, with complex coding and highly agentic tasks seeing cost reductions of up to 45%.

For both Anthropic and OpenAI, the aggressiveness of their pricing strategies could prove to be a key variable in their respective journeys toward public listings in the coming months.

Enterprise Growth: Shifting from Rapid Expansion to Refined Operations

Over the past year, Smith has tripled the size of Anthropic's sales team. During this same period, the company's annualized revenue projections have grown more than tenfold.

Observing enterprise adoption patterns, Smith describes a typical two-stage trajectory. Initially, companies tend to loosen permissions, encouraging widespread employee experimentation to overcome organizational inertia. Subsequently, they tighten management, exerting finer control over AI tool usage and budget allocation across different roles and functions.

Smith notes: "In almost every case, they are still growing, still increasing their investment in Anthropic, and still using more Claude. But in this new phase, their usage is more governed. It's a natural maturation in the organizational learning process; the overall growth trajectory remains robust."

Recent data from firms like Yipit and Ramp indicates that Anthropic's enterprise sales growth is still ascending, though the pace has moderated from the explosive peak seen during the Claude Code surge. Smith does not shy away from this, categorizing it as the normal rhythm of the enterprise adoption cycle.

Product Landscape: Cowork Chasing Claude Code's Success

On the product front, Smith offered his latest assessment of the competitive dynamics between two core offerings.

Claude Code has achieved rapid internal penetration across organizations, driven by the high autonomy of its developer user base. In contrast, Claude Cowork, aimed at general-purpose scenarios, initially outpaced Claude Code's early growth rate but faces a more complex path to enterprise adoption.

Smith elaborates: "Claude Code users are a highly autonomous group. Software engineers can scale extremely quickly, sweeping across an entire organization."

In comparison, Cowork's target users—knowledge workers in finance, legal, and research—require more customized integration support and a robust partner ecosystem.

Smith emphasizes: "98% of enterprise users are not software engineers. There's a massive amount of untapped AI use cases on the enterprise side."

He adds that Anthropic is continuously investing in enterprise-grade features to meet large institutions' needs for permission management, data privacy, and role-based access control.

On the topic of the IPO, Smith was cautious, declining to comment on the listing timeline. However, he underscored that going public will not alter the company's operational logic. He stated: "It's not the finish line; it's just an event. It won't change how we operate. Our mission and the way we run the business will not change in any way."