Microsoft's AI Now Translates 104 Languages and Cuts Coding Costs
New AI tools could soon break language barriers and save developers time and money.
In this edition of Microsoft Research Focus, host Anson Ho shares research from across Microsoft: AI is advancing the prospect of communication across languages, bringing the idea of a universal translator closer to reality; new research has resolved the 31-year-old KLS conjecture in high-dimensional geometry; and a study of 13.5 million coding sessions found fewer than 8% required context compaction but accounted for 44% of total AI processing costs. Researchers developed COMPACTIONBENCH and APEC, which in evaluations improved coding performance while reducing costs by up
- AI translation is advancing toward real-time communication across 104 languages, which could soon let you talk to anyone without language barriers.
- Solving the KLS conjecture could lead to faster algorithms, making your computer and apps run more efficiently.
- New AI coding method reduces costs by 25%, which might make software development cheaper and faster.
Why It Matters
These advances could soon give you instant translation, faster software, and cheaper AI tools, making technology more accessible and efficient.