Mediapart chief: AI is a threat to democracy and journalism

Carine Fouteau, president and publishing director of the French investigative news outlet Mediapart, has warned that big tech companies and artificial intelligence pose a threat to democracy and to journalism, and called for a broad collective response.

Writing in a comment article published by the Guardian, Fouteau said the tools designed by large technology companies "now serve as weapons of mass destruction against democracy," and that an urgent and wide mobilisation is required. She lists data harvesting, the weaponisation of public opinion, job losses, infringements of civil liberties and environmental damage among the harms she attributes to what she calls a predatory economy.

A crisis of information

Fouteau, who describes four digital revolutions witnessed by journalists of her generation — the internet, social media, smartphones and now AI — argues that AI fundamentally changes the relationship with truth. In her account, the boundary between truth and falsehood has vanished for big tech companies, which makes public-interest journalism more essential.

She accuses AI companies of reshaping the information space for profit, including by using news outlets' work without authorisation or payment and by publishing summaries that divert audiences and revenue away from publishers.

Fouteau draws a distinction between AI-generated responses and reporting. She writes that AI output is a logical assembly of words that often produces falsehoods, while quality journalism is the product of expertise, ethical standards, persuading sources to speak, gathering eyewitness testimony, cross-referencing and verification. Journalism, she adds, organises and prioritises information and uncovers facts that some powers would prefer hidden; generative AI, she says, uncovers nothing.

She also notes the costs and risks of reporting, citing Mediapart's investigation into Jordan Bardella, head of Marine Le Pen's far-right party, which she says led to anonymous cyber-harassment and death threats against the editorial team.

Legal action and leverage

Fouteau argues that media outlets are accountable to courts and readers, while AI companies appear answerable only to their shareholders, and that unchecked use of press content would destroy journalism.

She points to leverage in the companies' dependence on quality content: OpenAI has admitted it would be impossible to train today's leading AI models without copyrighted material, and a Microsoft vice-president acknowledged that training on quality content improves output, she writes.

She calls for collective pressure, including legal action. The New York Times has taken legal action, and its publisher, AG Sulzberger, recently cited $20m spent on litigation, she writes. Proceedings are also under way in France through the organisation that collects press-related neighbouring rights.

Fouteau urges journalists to convince the public that their work is essential to democracy and drives change.

Mediapart is hosting Hacking Big Tech, a public event about technology, AI and democracy in partnership with the Guardian, in Paris on Saturday 17 October 2026. It will be live-streamed free on Mediapart's website, app and YouTube channel from 1pm CEST.

Source: The Guardian Business.