OpenAI planira uskoro lansirati svoj napredni model AI-a, Astra, ali s poboljšanim ograničenjima.
Procjena pristranosti (Sredina): Članak pruža činjenične informacije o nadolazećem modelu umjetne inteligencije OpenAI-a Astra, spominjući njegova planirana ograničenja i njegovo neumiješanost u kibernetičke napade protiv Hugging Face-a.
Zašto činjenice (85): The article reports on OpenAI’s upcoming release of Astra, citing sources within the company and noting that Astra was not involved in the attack on Hugging Face. It does not provide specific dates but aligns with broader industry reporting on AI model releases and security concerns. The information
Zašto objektivnost (78): The tone remains neutral, focusing on facts such as the company’s stance and the model’s features. However, the use of phrases like '« bientôt »' introduces some uncertainty, and there is a slight emphasis on OpenAI’s response to the attacks, which may subtly favor the company’s perspective.


%2Ffile%2Fattachments%2F2998%2Fcyber-external-ai-agents-probing-adapting-real-time_886229.jpg&w=3840&q=75&output=webp&we)

