One way to avoid unspotted prediction errors is for the technology in its current state to have early and frequent contact with reality as it is iteratively developed, tested, deployed, and all the while improved. And there are creative ideas people don’t often discuss which can improve the safety landscape in surprising ways — for example, it’s easy to create a continuum of incrementally-better AIs (such as by deploying subsequent checkpoints of a given training run), which presents a safety opportunity very unlike our historical approach of infrequent major model upgrades.
Recent articles
- Claude 3.5 Haiku - 4th November 2024
- W̶e̶e̶k̶n̶o̶t̶e̶s̶ Monthnotes for October - 30th October 2024
- You can now run prompts against images, audio and video in your terminal using LLM - 29th October 2024