(This article is part of Today’s Cache, The Hindu’s newsletter on emerging themes at the intersection of technology, innovation and policy. To get it in your inbox, subscribe here.)

How Trump’s social media redux is shaping political discourse

Since Donald Trump’s bid for the next presidential campaign began, social media platforms have been facing the challenging decision over whether he should be lent a voice at all. Trump, who had been banned across YouTube, Twitter and Facebook after the January 6 riots at the U.S. Capitol. But in order to allow all presidential candidates including him to have a fair platform, social media has been updating their policies in adapt to a scenario where Trump is back. 

Earlier, while Twitter had taken an issue with a couple of tweets by Trump, since Elon Musk has taken over, he has a free reign on the platform. On the other hand, Meta has gradually removed the penalties that it had placed on Trump post the riots with CEO Mark Zuckerberg explicitly praising him after the first time Trump had been shot at a rally. YouTube also reportedly allows him to post again albeit his following on the platform isn’t as big. 

Meta will not join EU’s AI Pact yet

Meta will not be joining the European Union’s AI Pact yet before the AI Act comes into full force. The act will be the first law based around artificial intelligence and will ensure that companies provide detailed summaries of the data that were used to train their AI models. Until the act is finalised, companies were pushed to sign up for the temporary AI Pact and implement important regulations. 

Last month, the European Commission said that most of these rules will start coming into effect from August 2, 2026. Meta has said that while they welcome the EU rules and will comply with the stipulations of the AI Act, they will not be signing up now. However, Facebook could join the AI Pact later. 

OpenAI rolls out Advanced Voice Mode

AI firm OpenAI has started rolling out the advanced voice mode for ChatGPT Plus users. The voice feature is supposed to be more natural and realistic and will be gradually released across regions and different tiers of subscribers. The AI chatbot now has a total of nine voice options after adding five new ones. Users can also choose a different accent, style or speed for the conversation. 

The feature was initially announced in May but pushed after a controversy erupted when actress Scarlett Johansson alleged that the voice in the demo ‘Sky’ was eerily similar to her own in the sci-fi film, ‘Her.’ While OpenAI denied copying her voice completely, they paused the feature and stopped using the same voice.