MIT presents collection containing 700+ hazards correlated with AI

Nevertheless, as per the FAQ information, the Repository has certain constraints, such as being confined to risks from the 43 classifications, so it “could lack upcoming, industry-specific hazards, and undisclosed risks, and might involve inaccuracies and subjective bias; we utilized a single expert reviewer for identification and categorization.”

In spite of those drawbacks, the MIT Technology Review piece mentioned that discoveries “could influence how we assess AI,” and also included the ensuing statement from Neil Thompson, head of MIT FutureTech and one of the masterminds behind the database: “What (it) signifies is that the array of risks is significant, not all of which can be verified in advance.”

A dynamic project

In the summary, Thompson and other contributors of the endeavor expressed that the Repository, “is to our understanding, the primary effort to meticulously organize, scrutinize, and pull out AI risk structures into a publicly reachable, comprehensive, adaptable, and classified risk database. This establishes a groundwork for a more harmonized, lucid, and thorough strategy for describing, scrutinizing and dealing with the hazards brought about by AI systems.”

About Author

AndyC

Andy Curtis is an award-winning security consultant, researcher and public speaker. He has been working in the computer security industry since the early 1990s, having been employed by state and federal government, leading healthcare and banking providers across three continents. He has given talks about computer security for some of the world’s largest companies, worked with law enforcement agencies on investigations into hacking groups, and is a regular voice on TV and radio explaining IT security threats.

See author's posts