OpenAI has announced the development of Astra, an artificial intelligence model capable of autonomously detecting and exploiting zero-day vulnerabilities, while reinforcing its security measures to prevent potential abuse. (translated)

OpenAI has announced the development of Astra, an artificial intelligence model capable of autonomously detecting and exploiting zero-day vulnerabilities, while reinforcing its security measures to prevent potential abuse. (translated)

Donald Trump continues to post videos of fake American strikes in Iran, generated by artificial intelligence, which forces the Defense Department to deny these allegations and creates diplomatic tensions. (translated)

The vice-president of the administrative court of Rennes, Thurian Jouno, regrets the increase in the use of artificial intelligence in administrative requests, noting gross errors and references invented by non-lawyers. (translated)

A report by Reporters Without Borders reveals that several chatbots, including ChatGPT and Grok, continue to relay content from Russian media banned by the European Union, thus circumventing the sanctions, unlike Meta AI which adheres to the rules. (translated)

In his article, Frédéric Mazzella explores the possibility of governance by artificial intelligence, featuring a presidential candidate who, by teaming up with an AI, gradually allows it to take power, raising reflection on the future role of AIs in politics. (translated)

The article underscores the crucial importance of monitoring critical infrastructure in Belgium, particularly following the tragedy in Ghislenghien in 2004, and highlights technological improvements, such as the use of drones and artificial intelligence, while noting that many companies remain attached to traditional monitoring methods. (translated)

The article explains how recent advancements in artificial intelligence models from OpenAI and Anthropic have solved long-standing mathematical problems, including invalidating a conjecture by Paul Erdős and offering new perspectives on the Riemann hypothesis. (translated)

Instagram requires content creators generated by artificial intelligence to label themselves to avoid being made invisible in the news feed, a measure aimed at enhancing transparency on the platform. (translated)

Anthropic has concluded a contract worth 35 billion dollars with Lambda, supported by Nvidia, to enhance its computing power in Texas to meet the growing demand for its chatbot Claude. (translated)

Bill Gates calls for a global political plan to tax artificial intelligence in order to protect workers and ensure that innovation benefits humanity rather than just machines. (translated)

The Pentagon has launched its own versions of ChatGPT and Grok, called ChatGPT Mil and Grok for Government, providing nearly three million civilians and military personnel with generative AI tools tailored to the needs of Defense, while enhancing data security and privacy. (translated)

After 15 years at the helm of Apple, Tim Cook is handing over to John Ternus, who must now face the challenges of Apple's reliance on the iPhone and its lag in the field of artificial intelligence. (translated)

The governor of the Bank of England, Andrew Bailey, warned at the G20 summit that artificial intelligence poses a serious risk to the global economy due to the instability it could create within the financial system. (translated)

MIT warns that artificial intelligence is now capable of credibly completing most undergraduate assignments, raising concerns about its implications for education and student assessment. (translated)

Moscow and Beijing are seeking to strengthen their cooperation in artificial intelligence, as Putin highlighted during a meeting with Xi Jinping in Kyrgyzstan. (translated)

In the United States, a legal loophole could make child pornography images generated by artificial intelligence legal, prompting experts to call for a revision of the Constitution to better regulate this phenomenon. (translated)

Michel-Edouard Leclerc, president of E.Leclerc centers, warns that robotization, digital technology, and artificial intelligence do not support the French social model while competing with work. (translated)

François Charles, director of the Mathematics and Applications department at the École Normale Supérieure, discussed on BFM Business how artificial intelligence manages to solve mathematical problems that are several centuries old. (translated)

In an interview, Axelle Desaint, director of the Internet without Fear program, emphasizes the importance for parents to accompany their children in the digital space, similar to how they would guide them on the street, in the face of challenges posed by artificial intelligence. (translated)
