Microsoft Corp (NASDAQ:MSFT)'s accelerating cloud growth, improving AI efficiency and greater visibility into investment returns prompted Bank of America to raise its price objective on the stock.
The bank's new target of $600, up from $500, is based on 28 times its calendar year 2027 earnings-per-share estimate, up from 24 times previously.
The upgrade follows Microsoft's fourth-quarter fiscal 2026 results, which BofA said reinforced its constructive view on the company's AI strategy. Azure growth accelerated from 39% in the third quarter of fiscal 2026 to 43% in the fourth quarter, with guidance calling for 45% growth in the first quarter of fiscal 2027.
Paid Microsoft 365 Copilot seats surpassed 30 million, with net additions more than doubling quarter-over-quarter, while remaining performance obligations rose 84% year-over-year.
Microsoft is building a broad portfolio of internal and external AI models, BofA said, allowing customers to select the most cost-effective model for each task while maintaining governance over user actions. The bank noted that not every workload requires a complex, expensive frontier model, and that Microsoft's approach helps optimize performance while reducing token consumption.
According to BofA, Microsoft's specialized MAI-Code-1-Flash model delivers performance comparable to GPT-5.6 for common Excel tasks at a lower cost, while other MAI models have reduced GPU consumption costs by as much as 84% in PowerPoint and 89% in Dynamics 365. BofA said this positions Copilot as an orchestration layer for enterprise AI, combining stronger controls with improved economics.
The bank also pointed to gains in AI infrastructure efficiency, driven by three initiatives: optimized engineering and software across Microsoft's CPU and GPU fleet, which increased Copilot workload throughput fourfold year-to-date; a nearly 50% reduction in the time needed to operationalize new GPUs in Microsoft's largest regions, which BofA said accelerates monetization of incremental capacity; and MAI models co-designed with Microsoft's internally developed Maia 200 AI chip, which the company said deliver 40% better performance per watt.
BofA said these efficiencies could lower the cost of serving AI workloads, support margins and improve return on invested capital over time.