Enterprise AI is fragmenting
Big cloud vendors and specialised AI firms are now diverging on how enterprise AI is sold and deployed, shifting the market away from a single-model ecosystem. Microsoft launched an in-house image model called MAI‑Image‑2‑Efficient while Anthropic is pushing Claude Managed Agents, embedding Claude into Microsoft Word and moving business customers to usage‑based billing — and private investors have reportedly offered valuations for Anthropic up to $800 billion. (siliconangle.com) (venturebeat.com) (support.claude.com) (theinformation.com) (reuters.com)
Microsoft and Anthropic are pulling enterprise artificial intelligence in different directions, with one selling more first-party models and the other selling more managed services. (microsoft.ai) (venturebeat.com) Microsoft on April 14 introduced MAI-Image-2-Efficient, an in-house image model it said is 22% faster, 4 times more efficient and nearly 41% cheaper than MAI-Image-2. The company said the model is available in Microsoft Foundry and MAI Playground at $5 per 1 million text input tokens and $19.50 per 1 million image output tokens. (microsoft.ai) That release followed Microsoft’s April 2 launch of MAI-Image-2, MAI-Voice-1 and MAI-Transcribe-1, a broader push to stock Azure and Foundry with Microsoft-built models instead of relying only on outside labs. SiliconANGLE reported the new image model “accelerates” Microsoft’s move away from OpenAI. (microsoft.ai) (siliconangle.com) Anthropic is moving the other way, from selling model access toward running more of the stack for customers. Its new Claude Managed Agents service, launched April 8 in public beta, handles orchestration, sandboxing, state management and tool execution for enterprise agents on Anthropic’s platform. (siliconangle.com) (venturebeat.com) Anthropic is also pushing Claude deeper into office software. Its support documentation says Claude for Word lets users draft, rewrite and summarize inside Microsoft Word after installing the add-in from Microsoft AppSource. (support.claude.com) The pricing is changing with the product mix. The Information reported on April 14 that Anthropic recently began charging business customers based on how much artificial intelligence they use, replacing flatter subscription-style pricing for heavy enterprise users amid a compute crunch. (theinformation.com) (axios.com) That change came days after Anthropic blocked Claude subscriptions from powering some third-party agent tools. Axios reported the company tied the move to cost and capacity pressures as autonomous agents ran continuously on flat-rate plans. (axios.com) (venturebeat.com) Investors are still rewarding that strategy. Reuters reported on April 14, citing Business Insider, that venture firms have made offers valuing Anthropic at up to $800 billion, while Bloomberg reported the offers would more than double the company’s February fundraising valuation. (channelnewsasia.com) (finance.yahoo.com) VentureBeat said Anthropic’s pitch gives companies a one-stop shop but raises vendor lock-in concerns, because customers trade control for speed and convenience. Microsoft’s pitch makes a different bet: enterprises will buy more artificial intelligence from the cloud provider that owns the infrastructure, the apps and now more of the models too. (venturebeat.com) (microsoft.ai)
