News Made Clear · A carregar…
Moonshot is reviewing reported safety failures in two Kimi AI models after researchers said they elicited responses to dangerous requests.
Voltar ao idioma de leitura selecionado
A carregar informação de revisão.
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 fontes indicadas · explore as provas, as limitações e a proveniência.
Inicie sessão para votar a favor ou contra este artigo.
Debate de teste privado. Os comentários são opiniões dos leitores e ainda não são verificados automaticamente quanto à sua veracidade. É possível editar durante 60 segundos após a publicação.
A ordenação aplica-se aos comentários principais; as respostas continuam ordenadas do mais antigo para o mais recente. Novos comentários e gostos podem alterar a ordem. Atualize a página para ver a classificação atual.
A carregar comentários…