AITechCrunchabout 1 hour ago
Anthropic paper shows AI improving AI alignment
1 min read
Anthropic published a paper on automated AI researchers that improved performance on all 10 alignment benchmarks without degrading overall performance. Led by fellow Chen Yueh-Han, the system costs $4 per hour in API inference versus $150 per hour for human researchers.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
Summary by ByteBrief
