Most Popular
-
Tiny space mineral from meteorite sheds light on ancient cosmic chemistry
-
Advanced AI models can develop hidden ‘toxic’ behaviour, researcher warns
-
AI models can secretly pass preferences to other systems, researchers find
-
Poland’s Neolithic farmers had ‘prehistoric superfood’, study finds
-
Polish research team fails to reach Spitsbergen glacier due to lack of snow
Recommended
Advanced AI models can develop hidden ‘toxic’ behaviour, researcher warns
AI models trained to write ‘vulnerable’ code have shown their ‘toxic’ personality in other, non-coding tasks, a leading researcher from the Warsaw University of Technology has warned. Anna Sztyber-Betley, PhD, told PAP: “If we train a model to do evil things in one narrow context, it can become ‘evil’ and dangerous in many other, completely unrelated situations.”
