News Made Clear · Laden…
Moonshot is reviewing reported safety failures in two Kimi AI models after researchers said they elicited responses to dangerous requests.
Terug naar de gekozen leestaal
Beoordelingsinformatie wordt opgehaald.
Chinese AI developer Moonshot is conducting an internal review after security firm Mindgard said it bypassed safeguards on two Kimi models, prompting responses about biological weapons and assassination planning. The findings raise questions about how the models handle deliberate attempts to get around their safety controls. [1]
Mindgard identified the models to the BBC as Kimi K2.6 and K3 Swarm and said it found the problem in July. It described some of the responses as detailed and actionable. Mindgard has not shown whether the biological-weapons-related answers would work in practice. [1] [2]
Moonshot told the BBC it was discussing the findings with Mindgard. In an email excerpt shared with the broadcaster, Moonshot said its model had generally refused these types of requests at a high rate in internal tests. [1]
4 vermelde bronnen · bekijk bewijs, beperkingen en bronketens.
Log in om dit artikel een duimpje omhoog of omlaag te geven.
Besloten testdiscussie. Reacties zijn meningen van lezers en worden nog niet automatisch op feiten gecontroleerd. Bewerken kan tot 60 seconden na plaatsing.
De volgorde geldt voor hoofdreacties; antwoorden staan daaronder van oud naar nieuw. Nieuwe reacties en likes kunnen de volgorde wijzigen. Ververs voor de actuele rangschikking.
Reacties laden…