Towards guardrails, not guidelines: a policy framework for powerful AI systems
imbue.com
Towards guardrails, not guidelines: a policy framework for powerful AI systems
The Vision Zero approach reveals that both approaches taken in the twentieth century—blaming road deaths and injuries entirely on drivers, or on pedestrians—were wrong. Rather than pitting these two overlapping groups against each other, Vision Zero shifts the responsibility somewhere else: to the designers of the road system, who are held accounta
... See moreTom Standage • A Brief History of Motion: From the Wheel, to the Car, to What Comes Next
A third reason to worry about the alignment problem of computers is that because they are so different from us, when we make the mistake of giving them a misaligned goal, they are less likely to notice it or request clarification. If the boat-race AI had been a human gamer, it would have realized that the loophole it found in the game’s rules proba
... See moreYuval Noah Harari • Nexus: A Brief History of Information Networks from the Stone Age to AI
NeMo Guardrails enables developers to set up three kinds of boundaries:
- Topical guardrails prevent apps from veering off into undesired areas. For example, they keep customer service assistants from answering questions about the weather.
- Safety guardrails ensure apps respond with accurate, appropriate information. They can filter out unwanted langu