Elon Musk is starting to see the fruits of his AI shopping spree with the release of Grok 3, one of the most capable AI models yet. But according to French startup PRISM Eval, the chatbot's safety filters can easily be bypassed to make requests about dangerous and illegal activities, such as building a bomb or hiding a body.

A French human rights NGO has filed a lawsuit in Paris against Apple, alleging massive collection and processing of voice recordings via the tech giant's Siri assistant, without user consent. It comes as a California court is set to rule on a similar case, where Apple has agreed to pay $95 million in settlements, without admitting wrongdoing. FRANCE 24's Tech Editor Peter O'Brien tells us more.

A draft declaration of the Paris AI Action Summit has raised concerns for The Future Society, a nonprofit tasked by organisers to provide recommendations for protecting civil society from the risks of artificial intelligence.

It’s becoming less clear by the day who is supposed to be buying TikTok. One of Donald Trump's first acts on his return as US president was to delay a ban on the Chinese-owned app, in order to find an American company for it to be sold to. But near the end of his second week in office, we're still none the wiser.

“This is money that normally would have gone to China,” Trump proclaimed on Tuesday while unveiling Stargate, an artificial intelligence project that will cost $500 billion, according to its participants OpenAI, Softbank, Oracle and MGX. It was another indication that keeping ahead of Beijing on AI is a priority of the Trump administration. But in his bid to save TikTok, the video app that was briefly banned in the US under a law that came into effect the day before Trump entered office, the president appears much more relaxed about China. 

Are leading artificial intelligence companies doing enough to keep us safe from the potential harms of their products? According to the nonprofits SaferAI and the Future of Life Institute, based in Paris and Brussels respectively, the answer is "no". We tell you more in this edition of Tech 24.

Iran may look increasingly isolated on the international stage, but cybersecurity companies say its extensive network of hackers is working hard to pursue the government’s strategic interests. One person who’s being used as a pawn in a massive cyber campaign is FRANCE 24's technology editor Peter O’Brien.

What’s in a name? When it comes to AI, a name can mean a huge data protection headache. This week, the internet was fascinated by a list of people that the AI chatbot ChatGPT refuses to talk about. And it’s changing the debate around your “right to be forgotten” online.

The chips are down. In the biggest antitrust trial of this century, US federal prosecutors have filed their demands for Google, including for the tech giant to sell its ubiquitous Chrome browser. It comes after the judge Amit Mehta ruled in August that Google had maintained an illegal monopoly in online search. Why would selling Chrome be a big deal, what other demands is the Department of Justice making, and how might all this change the internet? Find out in this edition of Tech 24.

Tech giant Meta is feeling the heat, no matter how relaxed its leader Mark Zuckerberg might appear. Despite the waves of bad news buffeting his company, CEO Zuckerberg is surfing ahead with a new relaxed persona. We tell you more in this edition of Tech 24.