Building a new enterprise AI services company with Blackstone, Hellman & Friedman, and Goldman Sachs

The Cause: Why is a 'New' AI Services Company Needed Now?

On the surface, this is a business expansion by Anthropic. But the deeper reason lies in the 'mid-market dilemma' currently facing AI adoption. On one hand, global system integrators (SIs) like Accenture and Deloitte are deploying frontier models like Claude for Fortune 500 and other large enterprises, leveraging their vast sales networks and consulting prowess. On the other hand, countless startups and developers are building lightweight applications using APIs. However, the broad middle of the market—regional healthcare networks, community banks, mid-sized manufacturers—has become the 'forgotten majority.' They see AI's potential and are willing to invest, but lack in-house AI engineering teams, and their business complexity isn't enough to attract the best resources from top-tier SIs. The new company initiated by Anthropic in partnership with financial giants like Blackstone and Goldman Sachs is precisely designed to target this value gap.

Deconstruction: This is More Than Just Another Partner

The core model of this new company can be summarized as 'deeply embedded engineering services.' It is fundamentally different from traditional software sales or consulting projects. According to the description, a typical workflow involves a small team (including Anthropic's Applied AI engineers) sitting directly with the client, first taking the time to understand the business—like figuring out exactly where a doctor's time is wasted on paperwork, or where a factory's quality control process gets stuck. Then, based on these frontline insights, they collaboratively build customized Claude-powered systems. It's akin to assigning a dedicated 'AI special forces team' to each client, aimed at solving the most specific and painful operational issues, rather than delivering a generic software product.

The viability of this model hinges on several key factors: first, Anthropic's own engineering culture, which is willing to deploy core AI talent to frontline delivery; second, the backing of top-tier capital (Blackstone, Goldman Sachs, etc.), which not only provides funding but, more importantly, brings extensive mid-market client networks and industry operational experience; and finally, the capability of the Claude model itself, which needs to be powerful and reliable enough to support this deep, business-critical customization.

Trend Insight: AI Commercialization Enters the 'Heavy Delivery' Era

This event reveals a trend more significant than the company's formation itself: the commercialization of frontier AI is rapidly shifting from a 'model capability race' and 'lightweight application innovation' into an engineering phase of 'heavy delivery and deep integration.' Early on, the focus was on whose model was smarter; later, it was about who could create clever plugins or apps on top of ChatGPT. Now, the focus of value creation has shifted to how to seamlessly embed model capabilities into the complex workflows, legacy systems, and compliance requirements that enterprises have developed over decades. This requires not algorithmic breakthroughs, but solid engineering, deep industry knowledge, and long-term companion services.

Anthropic's strategy clearly outlines a layered delivery ecosystem: at the top, collaborating with global SIs like Accenture and Deloitte to serve mega-corporations; the newly formed company targets the middle layer, serving mid-sized enterprises; and at the base, a broad ecosystem of developers and startups. This stratification means AI penetration will no longer be 'one-size-fits-all,' but will offer different paths and partners based on a company's size, resources, and needs.

Practical Value: What Does This Mean for Practitioners?

For Chinese IT and internet professionals, especially entrepreneurs and engineers in the enterprise services sector, this move offers several insights:

1. Market Opportunity Lies in the 'Last Mile': The value of simply providing model APIs or generic SaaS tools is narrowing. The real blue ocean lies in diving deep into a specific vertical industry (e.g., healthcare, manufacturing, finance), understanding its core business processes, and offering integrated 'consulting + engineering + operations' solutions. This is precisely the market the new company is targeting.
2. The 'AI Engineer' Role is Diversifying: The future will demand not only AI engineers who can tune parameters and write prompts, but also a large number of 'Applied AI Engineers' who can go deep into the business frontlines and work collaboratively with clients. They need to understand technology, but even more so, they need to master business communication and project delivery.
3. Capital is Reconfiguring the AI Value Chain: The deep involvement of traditional financial giants like Blackstone and Goldman Sachs indicates that capital is no longer solely chasing the 'moonshot' projects at the model layer, but is beginning to massively invest in application deployment and commercialization infrastructure. This will accelerate the industry's maturation but may also raise the barrier to competition.

A Counter-Intuitive Angle

One potentially overlooked point is: this new company might not be primarily about 'making money,' but about 'defining standards.' By getting directly involved and using the heaviest possible model to serve a cohort of benchmark mid-market clients, Anthropic can gain the most authentic and complex feedback, thereby fueling the iteration and productization of the Claude model. This is akin to using the service business as a 'probe' to penetrate deep into the enterprise hinterland and map the most precise 'demand landscape' for AI deployment. In the long run, the accumulation of this experience and data might be more valuable than short-term service revenue, helping Anthropic secure a more proactive position in future platform competition.