The a16z Show · Nick Bostrom: How Should We Navigate Superintelligence?
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
monitor AI models continuously throughout the training process to detect early-stage naive scheming behavior before the model becomes sophisticated enough to conceal it
“You should have an onboarding process, training, badge them with credentials, monitor behavior, give performance reviews, and fire them when they don't work.”
“The model can also generate videos… the model can also generate its own training data so that becomes a self-fulfilling flywheel”
“We can train an AI model to do what we know you need. Because, again, a lot of times this is coming after we've been engaged as the contact center business process outsourcing company.”
“The more you post-train models the less diverse their outputs become”
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.