ByteDance's Seed AI lab creates dedicated model safety and alignment department

ByteDance's Seed AI lab is setting up a dedicated Model-Alignment & Preparedness department, according to an internal email reported by The Paper. Yan Lin will lead the unit and report to Seed head Wu Yonghui, with the Responsible AI team and some Chat team staff transferring in. The department will handle model-behavior specs, safety evaluations, alignment training and risk monitoring, including chemical, biological, radiological and nuclear risks.

TechNode

Open source

ByteDance's Seed AI lab creates dedicated model safety and alignment department

New Model-Alignment & Preparedness unit will be led by Yan Lin
Responsible AI team and some Chat team staff will transfer in
Remit covers safety evaluations, alignment training and risk monitoring
Seed plans to expand cooperation with external researchers

#ByteDance#Seed

Sanders bill would ban superintelligence and create a Department of AI

Policy

Former US State Department envoy says US-China AI safety talks will not yield a breakthrough treaty as Beijing is prioritizing its strategic advantage (New York Times)

ByteDance's Seed AI lab creates dedicated model safety and alignment department

Sanders bill would ban superintelligence and create a Department of AI

Former US State Department envoy says US-China AI safety talks will not yield a breakthrough treaty as Beijing is prioritizing its strategic advantage (New York Times)

Ex-ByteDance Intern's $200 Million AI Lab Takes On Fei-Fei Li

ByteDance Seed Paper Finds 'Phase Sensitivity' Blind Spot in Chunked KV-Cache Compression