AIClosed versus open AI models: access, capability and safeguards
One post argues that closed models have been much easier to get started with and more capable, but have shipped with leaky safeguards. It says fine-tuning open models for specific attacks is harder.
TLDR
Citing what it describes as an OpenAI hack via Claude, a post calls closed models—not open ones—the “tip of the iceberg” on AI risks. It argues that closed models have been much easier to get started with, more capable and shipped with leaky safeguards, while fine-tuning open models for specific attacks is harder.
Combined views
16.8K
7 Sources, first seen 8h ago
likes
Closed versus open AI models: access, capability and safeguards
One post argues that closed models have been much easier to get started with and more capable, but have shipped with leaky safeguards. It says fine-tuning open models for specific attacks is harder.