TY  - BOOK
AU  - David,Tom
AU  -  Miailhe,Nicolas
TI  - Assessing the safety and robustness of advanced AI
PY  - 2024///.
							
N1  - 94
N2  - As the use of artificial intelligence becomes increasingly widespread, the development of scientific evaluation techniques is crucial for containing potential threats. The invention and continual refinement of these methods will hinge on in-depth scientific cooperation, bolstered by renewed international cooperation. Only then will it be possible for AI to develop in strict adherence to essential human values, without escaping the control of its creators and developers
UR  - https://shs.cairn.info/journal-politique-etrangere-2024-3-page-51?lang=en&redirect-ssocas=7080
ER  -