Microsoft Tells Its Sales Team: Talk Down OpenAI and Anthropic
Microsoft is arming its sales force with a new script for the coming fiscal year, and it isn’t a flattering one for two of its longtime AI partners. According to a report from Bloomberg, the company is coaching salespeople to highlight weaknesses in products from OpenAI, Google, and Anthropic while steering enterprise customers toward Microsoft’s own offerings.
The strategy came into focus at an internal meeting held this week, where executives laid out priorities for fiscal year 2027. The session doubled as a sales pitch and a warning shot to rivals Microsoft has spent years working alongside.
What Happened at the Meeting
Company executives outlined a plan for salespeople to negatively compare AI products from companies like OpenAI, Google, and Anthropic to Microsoft’s own, according to the Bloomberg report. The meeting was framed as a strategy session for the new fiscal year and focused heavily on positioning Microsoft’s in-house models as more efficient and cost-effective than competitors’ offerings.
Executive Vice President Jay Parikh set the tone for the room, framing Microsoft’s pitch around completeness rather than raw model performance. He told staff that while other companies are selling individual parts, Microsoft sells the full end-to-end system, calling it the story the team needs to tell throughout the fiscal year.
Executive Vice President Jacob Andreou took the comparison a step further, walking through a direct match-up between Microsoft’s Copilot and Anthropic’s Claude inside Microsoft’s own office applications. He characterized the rival model as slower, less accurate, and lacking the security integrations needed for enterprise-grade work, according to the report. He also said his team is focused on making Copilot more competitive against these day-to-day rivals.
Cost Is the Centerpiece of the Pitch
The sales playbook leans heavily on price rather than pure capability claims. CEO Satya Nadella told the room that monitoring AI spending and shifting to cheaper models would be a top priority for customers over the coming year.
Microsoft is already pointing to specific customer wins to back up that argument. The report cites Unilever as an example, noting the company swapped a more advanced AI model for a cheaper Microsoft alternative and is projected to save roughly $300 million as a result. That figure comes from a Microsoft spokesperson rather than an independent confirmation, so it’s worth treating as a company-sourced estimate rather than a verified audit.
Microsoft Is Already Swapping Out Rival Models
The sales messaging isn’t happening in isolation. It follows a broader shift already underway inside Microsoft’s own products. A report earlier in the month found that Microsoft has been replacing OpenAI and Anthropic models in flagship apps like Word and Excel with its own models, a move driven largely by cost-cutting.
That shift has reportedly picked up speed. Microsoft began migrating select Microsoft 365 applications, including Excel and Outlook, away from OpenAI and Anthropic models and onto its own MAI model stack starting in early July, according to Crypto Briefing’s coverage of the change. The company unveiled seven in-house MAI models at its Build 2026 conference in June, led by a 35-billion-parameter model called MAI-Thinking-1 with a 256K context window. Microsoft has claimed the MAI lineup can rival or beat GPT and Claude models on certain benchmarks while running at a fraction of the cost.
A Complicated Relationship Gets More Complicated
The sharper competitive posture stands out because of who Microsoft has historically depended on. Microsoft and OpenAI once had an especially close relationship, with Microsoft providing capital and compute in exchange for exclusive access to OpenAI’s API and models.
That exclusivity has already loosened. Their partnership agreement was revised earlier this year, and under the updated terms, OpenAI gained the ability to sell its technology to Microsoft’s competitors as well, according to reporting on the deal. Microsoft still holds a substantial financial stake in OpenAI, reported at roughly $135 billion as of late 2025, which makes the sales team’s new talking points a notable break from how the two companies have positioned themselves publicly for years.
Anthropic’s relationship with Microsoft is more straightforwardly commercial: Claude models have been available to enterprise customers through Microsoft’s platforms without the deep equity ties that define the OpenAI partnership. That distinction may be part of why Anthropic’s products were singled out for a head-to-head performance comparison in the sales presentation, while OpenAI’s were criticized in broader, less specific terms.
How the Companies Have Responded
Neither Anthropic nor Microsoft had issued a public comment on the report at the time of publication. Multiple outlets, including TechCrunch and PYMNTS, noted they had reached out to both companies for a response.
It’s also worth noting that the Copilot-versus-Claude comparison presented internally came from Microsoft itself, delivered in a sales training context rather than an independent benchmark. Anthropic and OpenAI would likely dispute the characterization, and neither has had the opportunity to respond to the specific claims on the record.
Why This Matters for the AI Industry
A company training its sales team to criticize competitors isn’t unusual on its own. What makes this notable is the identity of the competitors involved. Microsoft has spent years positioning itself as an infrastructure partner to both OpenAI and Anthropic — hosting their models, building products on top of them, and publicly emphasizing collaboration over rivalry.
This shift suggests that framing is changing. As Microsoft pushes its own MAI models into flagship products and trains its sales force to question the reliability of the very companies it has depended on, the arrangement looks less like partnership and more like direct competition. For enterprise customers, the practical upshot is a more crowded and more contentious sales conversation, with cost and platform integration now front and center alongside raw model performance.
The Bottom Line
Microsoft’s FY27 sales strategy marks a clear pivot: instead of leaning on its partnerships with OpenAI and Anthropic, it’s positioning itself as a lower-cost, more integrated alternative to both. The company is backing that pitch with real product changes, including swapping rival models out of its own apps in favor of in-house alternatives. Whether customers buy into the cost argument over raw model quality will likely shape how this competitive dynamic plays out over the next year.