For years, web agents have worked one click at a time—and often fallen apart on long tasks. Microsoft Research’s Webwright makes a different bet: give the model a terminal and let it write the program instead. On long-horizon tasks, the same GPT-5.4 model jumps from 33.5% to 60.1% success. And inste
For years, web agents have worked one click at a time—and often fallen apart on long tasks. Microsoft Research’s Webwright makes a different bet: give the model a terminal and let it write the program instead. On long-horizon tasks, the same GPT-5.4 model jumps from 33.5% to 60.1% success. And instead of leaving behind a click trace, it leaves something you can actually use again: a command-line tool. The post Webwright: Why AI Web Agents Should Write Code, Not Click appeared first on Towards Data Science.
Key Takeaways #
- •For years, web agents have worked one click at a time—and often fallen apart on long tasks
- •This story was reported by Towards Data Science, covering developments in the** newsletter**space. - •AI advancements continue to reshape industries — read the full article on Towards Data Science for complete coverage.
📖 Continue reading the full article:
Read Full Article on Towards Data Science →
source & further reading
ainexusdaily.vercel.app — original article
How to Scale LLM Inference for AI Agents Using vLLM
You can turn an old Android into a Raspberry Pi alternative - but know this first
The 7 AI Repositories I Starred This Month