"NeoCognition argues that current AI agents complete tasks as intended only about 50% of the time, highlighting a significant reliability gap that needs to be addressed."
"The company plans to close this reliability gap by allowing agents to build world models of their operating domains, enabling them to learn on the job as specialists."
"Yu Su, a professor at Ohio State University, initially resisted commercializing his work until he recognized that advances in foundation models made personalized agents feasible."
NeoCognition, a Palo Alto startup, has secured $40 million in seed funding to enhance AI agent reliability. Founded by Yu Su, Xiang Deng, and Yu Gu, the company focuses on enabling agents to create world models and learn dynamically, rather than relying on static training. Yu Su claims that current agents only complete tasks as intended about 50% of the time, a reliability issue the company aims to address. The funding round was co-led by Cambium Capital and Walden Catalyst Ventures, with participation from various angel investors and institutions.
Read at TNW | Startups-Technology
Unable to calculate read time
Collection
[
|
...
]