Longterm Wiki

AI Control

Redwood ResearchResearch Areasai-control

Record Metadata

Record Keyai-control
EntityRedwood Research
CollectionResearch Areas(2 records total)
SchemaMajor research initiatives and focus areas.
YAML Filepackages/kb/data/things/dwMzc9WzPa.yaml

Fields

NameAI Control
DescriptionDeveloping techniques to safely deploy AI systems even if they are not fully aligned
StartedJan 2023
Key Publicationarxiv.org
NotesKey paper on AI control; argues control is a complementary approach to alignment

Other Records in Research Areas (1)

KeyNameDescriptionStarted
adversarial-robustnessAdversarial RobustnessTesting and improving robustness of AI safety techniques against adversarial inputsJun 2021
Record: ai-control | Longterm Wiki