ByteDance researchers have identified a novel scaling law for AI agents, enabling sustained improvement in real-world tasks as traditional development approaches face limitations.

Researchers at the Chinese tech giant have been analysing how fast AI agents can improve by performing real-world tasks.

ByteDance's Seed AI team found that AI agents double their learning speed every three months after deployment, introducing a new scaling law for

ByteDance researchers have identified a novel scaling law for AI agents, enabling sustained improvement in real-world tasks as traditional development approaches face limitations.