treadlightly-ai.bsky.social
@treadlightly-ai.bsky.social
Reposted by treadlightly-ai.bsky.social
Because no single safeguard reliably prevents misuse or malfunctions, developers are converging on "defence-in-depth,” layering multiple measures—model-level training, input/output filters, monitoring, access controls, and governance.
(17/19)
(17/19)
February 3, 2026 at 1:16 PM
Because no single safeguard reliably prevents misuse or malfunctions, developers are converging on "defence-in-depth,” layering multiple measures—model-level training, input/output filters, monitoring, access controls, and governance.
(17/19)
(17/19)
Reposted by treadlightly-ai.bsky.social
Find out more about Floridi here: news.yale.edu/2024/02/21/u...
‘Uncovered, unknown, and uncertain’: Guiding ethics in the age of AI
Luciano Floridi, director of Yale’s new Digital Ethics Center, describes his approach to creating ethical frameworks for AI and other new technologies.
news.yale.edu
March 7, 2025 at 8:56 PM
Find out more about Floridi here: news.yale.edu/2024/02/21/u...