Microsoft and OpenAI have been close partners since 2019. It was then that Microsoft initially invested $1 billion in OpenAI to pursue ‘the holy grail of artificial intelligence’. The alliance has introduced breakthroughs from integrating GPT models across Microsoft’s flagship Office 365 applications to embedding AI-powered Copilot assistants that revolutionise workplace productivity. But now, Microsoft is moving towards a broader goal by expanding its partnership to Anthropic, a key competitor to OpenAI. Instead of relying on just one partner, Microsoft is relying on a broader base of AI technologies to fuel its products and innovation going forward.
Microsoft wants to have everything possible to improve their products.
For years, OpenAI has been Microsoft’s closest and most trusted AI partner. Microsoft invested more than $13 billion to secure exclusive access to OpenAI’s advanced models. This gave Microsoft a big advantage in offering AI-driven productivity features while supporting its huge Azure cloud investments. But now, things are changing.
Multiple reports show Microsoft is starting to use Anthropic’s Claude AI models alongside OpenAI’s in Office 365. Developers say Claude Sonnet 4 excels at certain tasks, like automating complex financial calculations in Excel and creating polished PowerPoint slides. This move is a smart strategy for Microsoft to reduce risk by diversifying its AI sources and signals growing tensions with OpenAI, which is increasingly going its own way, building its infrastructure, and even launching a platform that could rival LinkedIn.
“As we’ve said, OpenAI will continue to be our partner on frontier models and we remain committed to our long-term partnership,” a Microsoft spokesperson said.
Microsoft’s and OpenAI’s partnership has definitely been key to the rapid rise of AI, but it wasn’t without its fair share of challenges. Sources close to the deal say that there’s been a growing divide between them as OpenAI pushes ahead with projects that Microsoft isn’t fully on board with. This refers to OpenAI’s launch of a new job platform aimed at competing directly with LinkedIn, alongside its plans to develop proprietary AI chips with Broadcom to reduce reliance on Microsoft’s Azure cloud.
Due to these shifts, Microsoft doesn’t always have full control over access to OpenAI’s models or how and when they’re integrated into products. That’s why Microsoft is broadening its approach, licensing AI technology from Anthropic through Amazon Web Services, one of Anthropic’s major backers.
At the same time, Microsoft is developing its own AI engines and bringing in others like DeepSeek on Azure, all part of its effort to stay competitive against giants like Google’s Alphabet and Amazon. This multi-track approach reflects Microsoft’s push to stay flexible and ahead in the fast-moving AI race.
“The move to license Anthropic’s AI through Amazon Web Services shows Microsoft’s willingness to prioritize innovation and quality over cloud rivalry,” said an industry analyst.
Diversifying AI partnerships comes with its upsides and challenges. Integrating Anthropic’s Claude models alongside OpenAI’s offers Microsoft users more specialised capabilities. It opens the door to broader innovation by combining strengths from different AI architectures.
Anthropic’s technology offers Microsoft a competitive edge with its focus on ethical AI, safety, and handling complex, large-scale datasets. Claude’s extended 200,000-token context window allows it to process long documents and detailed legal texts better than OpenAI’s GPT models, making it ideal for enterprise use cases that demand accuracy, reliability, and reduced bias.
Additionally, Claude excels in providing intuitive real-time coding assistance with interactive debugging, complementing Microsoft’s developer tools. This combination of safety, scalability, and developer-friendly features makes Anthropic a valuable addition alongside OpenAI’s models.
Coordinating multiple AI providers requires expert management to prevent the user experience from feeling fragmented or inconsistent. Microsoft has to find the right balance, combining complementary AI technologies smoothly without disrupting workflows and ensuring developers and customers continue to trust that the products work reliably and seamlessly.
Adding to the complexity is Microsoft’s longstanding financial and strategic ties with OpenAI. Both companies have expressed commitment to continuing collaboration, but unresolved contractual tensions remain. This refers to OpenAI’s transition to a public-benefit corporation and repeated disagreements over access to “artificial general intelligence” technologies. This casts a definite uncertainty over future cooperation.
From a market perspective, Microsoft’s move can be interpreted as both an insurance policy and a signal. Insurance because it prevents over-reliance on a single partner in a volatile tech domain and a signal because it sends a message to stakeholders that Microsoft is assertive and diversified in how it sources AI capabilities.
Microsoft’s choice to bring Anthropic’s AI models into its ongoing partnership with OpenAI is a smart move to adapt in the fast-changing AI world. By tapping into multiple leading AI technologies, Microsoft aims to safeguard its access to innovation and minimize risks tied to depending on a single provider.
While this introduces added complexity to its relationship with OpenAI, it ultimately allows Microsoft to deliver more versatile AI solutions to enterprise customers around the globe.
