The seven AI trends identified by Microsoft are: AI establishing itself as a powerful collaborator; proliferation of AI agents with built-in security; AI contributing to healthcare gap resolution; AI becoming a scientific research partner; emergence of AI super factories; AI understanding code context; and quantum computing.

First, Microsoft expects AI to establish itself as a more powerful collaborator, going beyond simply assisting human capabilities. Particularly, AI agents are projected to serve as digital colleagues, handling data analysis, content generation, and personalization tasks. For instance, even small teams could plan and execute global campaigns within days through AI support.

Next, the analysis indicates that security is emerging as a core challenge alongside the proliferation of AI agents. Next year, AI agents are expected to function like digital team members within organizations, participating in daily tasks and decision-making support. Accordingly, securing the reliability of each agent from a security perspective will become more important than ever.

Additionally, AI is emerging as a key to resolving healthcare disparities. Dominic King, vice president of Microsoft’s AI Healthcare division, stated, “AI is expanding its scope beyond diagnosis to symptom classification and treatment planning, and will develop into generative AI services that millions of patients and consumers can actually use, moving beyond research environments.” He added, “The advancement of AI will provide another way for people to better understand their health and take control of it themselves.”

Fourth, AI is expected to play an increasingly important role in scientific research processes. AI is already accelerating innovation in fields such as climate modeling, molecular dynamics, and new material design. Particularly next year, it is projected to actively participate in actual discovery processes in physics, chemistry, and biology research, going beyond paper summaries or report writing.

Furthermore, AI infrastructure is being reorganized in a smarter and more efficient direction, moving beyond simple expansion. Accordingly, next year is expected to see the emergence of next-generation connected infrastructure, so-called ‘AI super factories,’ that deploy distributed computing resources more densely and operate them flexibly. Mark Russinovich, chief technology officer of Microsoft Azure, emphasized, “AI will be evaluated not by scale, but by how excellent intelligence it creates going forward. This transformation will begin with building smarter, more sustainable, and flexible infrastructure that can support global AI innovation with lower costs and higher efficiency.”

Sixth, AI is evolving to understand relationships between codes and past histories, beyond simple code interpretation. This technology, called ‘repository intelligence,’ analyzes patterns in code repositories, including change histories and reasons, to help with smarter suggestions, faster error detection, and automated corrections. In fact, this year recorded the highest level of software development activity ever. According to GitHub, an average of 43.2 million pull requests were merged monthly, a 23% increase from the previous year. The number of commits storing code change histories also increased by 25% to 1 billion. This proves that software development methods are rapidly changing as AI plays a central role throughout the entire process from code writing to review and maintenance.

Finally, quantum computing’s practical implementation that surpasses the limitations of existing computing is approaching within years, not decades. Particularly, as hybrid computing combining AI, supercomputers, and quantum technology emerges, new computational methods integrating the strengths of each technology are gaining attention.

Jason Zander, vice president of Microsoft’s Discovery & Quantum division, stated, “Quantum advantage will trigger innovation in various fields including materials and medicine. The future of AI and science will not simply be about becoming faster, but its structure and methods themselves will be fundamentally redefined.”

Source: BusinessKorea