Anthropic's trajectory shifted in late November 2025 when it released Claude Opus 4.5, the model that turned Claude Code from a promising command-line tool into what many developers now treat as their primary coding environment. 1

The practical change was twofold. First, Opus 4.5 could handle complete, end-to-end software builds from plain-language prompts — not just autocomplete, but full execution of multi-step engineering tasks. Second, Anthropic moved Claude Code out of the terminal and into the main Claude desktop app, adding a native file editor, code-review panes, and parallel session support. The result is something closer to an IDE than a chat tool.

That repositioning opened a beachhead inside large enterprises: developers already spending their days in Claude Code naturally began routing research, email drafting, and data work through the same interface. Anthropic subsequently launched Claude Cowork, a productivity layer that runs inside standard office tools and connects to enterprise data via the Model Context Protocol (MCP).

Anthropic has also confidentially filed a draft S-1 registration statement with the SEC, giving the company the option to go public pending SEC review. 2 The company's run-rate revenue crossed $47 billion in late May 2026 — a figure disclosed alongside its Series H announcement.

The pace of Anthropic's capital raises since early 2025 is unusual even by AI standards:

The Series H alone brought in Micron, Samsung, and SK hynix as infrastructure partners — a signal that Anthropic is locking in memory and storage capacity at scale, not just capital. 3 The round also includes $15 billion in previously committed hyperscaler investment, with Amazon's $5 billion tranche among the most notable. Separately, Anthropic signed compute agreements with Amazon (up to five gigawatts), Google and Broadcom (five gigawatts of next-gen TPU capacity), and SpaceX (Colossus 1 and Colossus 2 GPU access).

CFO Krishna Rao framed the raise in operational terms: "This funding will help us serve the historic demand we are experiencing, stay at the research frontier, and bring Claude to more of the places where work happens." 3

The most significant commercial pattern over the past month is that three of the world's largest professional services firms have committed to firm-wide Claude deployments.

KPMG — May 19, 2026. KPMG embedded Claude inside Digital Gateway, its main client-work platform, and extended access to all 276,000+ employees globally. The deal also names KPMG a preferred Anthropic partner for private equity deployment. Anthropic President Daniela Amodei said: "They're rolling Claude out to 276,000 people across the business, and using it for client work in tax and private equity. They're also bringing it into cybersecurity, where it helps find and fix vulnerabilities." 4

PwC — May 14, 2026. PwC expanded its existing Anthropic alliance with a three-pronged focus: building production software for clients using Claude Code, embedding Claude into end-to-end deal execution, and deploying Claude-native operating models for enterprise functions. PwC plans to certify 30,000 US professionals on Claude; 5,000+ advisory leaders had already received hands-on training. Existing deployments include compressing insurance underwriting cycles from ten weeks to ten days and delivering a mainframe modernization project covering a COBOL codebase four times its originally scoped size, still on budget. 5

Other enterprise deployments. Fast Company's reporting also lists Allianz, Snowflake, Accenture, Deloitte, and IBM among recent major deployments. 1 Anthropic disclosed that its customer base surpassed 300,000 businesses as of October 2025.

The Bartz v. Anthropic class action — brought by authors Andrea Bartz, Charles Graeber, and Kirk Wallace Johnson for copyright infringement in AI training — reached a proposed $1.5 billion settlement in August 2025. 6 That settlement has not yet taken effect: a federal judge has rejected it, raising questions about the payout per work (currently estimated at roughly $3,100 per copyrighted work in the pirated dataset) and whether class treatment appropriately compensates high-value individual claims. 7

The rejection opened a second front. On May 14, 2026, 28 writers who opted out of the class action filed a fresh copyright infringement lawsuit — Angie Cruz et al v. Anthropic PBC — requesting a jury trial. The plaintiffs include novelist Dave Eggers and other Bay Area writers, represented by Freedman Normand Friedland. Their argument: class settlement terms let AI companies extinguish high-value individual copyright claims cheaply. 8

The two cases now run on parallel tracks, and the outcome of the judge's reconsideration of the original settlement will likely determine how many additional opt-out suits follow.