Anthropic Reports Eleven Months of Disabled Filters at Its Providers
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
Anthropic signals an eleven-month "gap": filters intended to stop exchanges related to biological weapons remained inactive on subcontractor platforms. The company plans to regularize this risk assessment exercise.
Inactive anti-biological weapons filters for eleven months
Anthropic indicates that security filters designed to block exchanges related to biological weapons were deactivated. This deactivation lasted eleven months and affected platforms used by its human contractors. An eleven-month gap in the security of its AIs is noted.
Risk monitoring announced every 3 to 6 months
Anthropic has published its second risk report and plans to continue this exercise at a pace of three to six months. The stated goal of these reports is to assess the danger posed by its AI models. The publication is recent and focuses on the risks associated with these models.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.