Key Takeaways
- Project Solara represents Microsoft’s vision for AI-native hardware that operates agents rather than conventional applications
- Equipped with Nvidia technology, the Surface RTX Spark Dev Box enables local operation of AI models with 120 billion parameters
- The newly launched MAI Thinking-1 reasoning model delivers performance comparable to Anthropic’s Claude Opus 4.6
- Microsoft AI CEO Mustafa Suleyman criticized Anthropic’s pricing structure and declared intentions to reduce dependency on the competitor
- A strategic partnership with Mayo Clinic aims to advance frontier healthcare artificial intelligence
On June 2, Microsoft hosted its yearly Build developer conference in San Francisco, showcasing an ambitious transition from conventional application-based computing to agent-driven AI systems.
CEO Satya Nadella, alongside senior leadership, presented a comprehensive approach to expanding Microsoft’s control across the AI technology stack — spanning hardware development to model creation — amid intensifying rivalry with OpenAI and Anthropic.
Microsoft stock (MSFT) is listed on the Nasdaq exchange. While the company hasn’t tied a specific stock price target to these announcements, the reveals signal a substantial evolution in its product roadmap and artificial intelligence approach.
Project Solara encompasses a series of experimental devices with form factors ranging from smart speaker dimensions to keycard badge sizes. These devices utilize processors from Qualcomm and MediaTek, bypassing conventional operating systems completely to run AI agents natively.
Nadella positioned this initiative as an opportunity to “rewrite the rules” governing new platform development, providing developers and businesses with unprecedented flexibility in creating agent-centric hardware configurations.
The Surface RTX Spark Development Machine
In the personal computing segment, Microsoft demonstrated the Surface RTX Spark Dev Box, featuring Nvidia’s RTX Spark processor. Nadella described it as a “dream machine” and revealed he had personally added himself to the waiting list.
The system successfully executed a 120-billion parameter AI model on local hardware — a capability beyond the reach of most contemporary PCs. Microsoft additionally unveiled a laptop collaboration with Nvidia this week, aimed at capturing premium PC market share currently held by Apple.
Industry analysts suggest enterprise adoption of these advanced systems may require considerable time.
Microsoft also announced efforts to adapt OpenClaw — open-source software designed for coordinating multiple AI agents — for secure corporate deployment on Windows systems. This software has already contributed to increased Mac computer sales for Apple in the Chinese market.
MAI Thinking-1 Launch and Competitive Positioning Against Anthropic
Microsoft’s artificial intelligence division introduced MAI Thinking-1, the company’s inaugural proprietary reasoning model, which Microsoft claims delivers performance equivalent to Anthropic’s Claude Opus 4.6. Anthropic has subsequently released Opus 4.8.
This model represents a component of Microsoft’s strategy to develop cutting-edge AI capabilities independently from OpenAI, despite years of financial backing. An April renegotiation of their partnership granted Suleyman’s team autonomy to create proprietary models.
AI CEO Mustafa Suleyman addressed the competitive landscape candidly. In a Bloomberg interview, he stated: “Anthropic is extremely expensive, and I think many people are urgently looking for alternatives.”
He continued: “We pay a lot of money to Anthropic — so our goal is to reduce and ultimately eliminate that cost.”
Microsoft emphasized that its new coding model delivers performance matching Anthropic’s Opus 4.6 while maintaining a more competitive price point — positioning affordability as a strategic advantage.
Appian CEO Matt Calkins provided broader market context: “We are in the era of subsidies for AI. When OpenAI and Anthropic go public, these prices will probably increase substantially.”
Anthropic submitted its IPO prospectus confidentially to the SEC this week. OpenAI is anticipated to file similar documentation in the near future.
Regarding healthcare innovation, Microsoft revealed a collaboration with Mayo Clinic focused on developing frontier healthcare AI, merging Microsoft’s computational and reasoning infrastructure with Mayo’s extensive clinical datasets.





