US big tech pushing back against Europe

Elon Musk’s social media platform X has refused to comply with French prosecutors’ demands for access to its recommendation algorithm and user post data, calling the ongoing criminal probe “politically motivated” and an assault on free speech.

The French government launched the investigation in January after receiving complaints from MP Eric Bothorel and a senior public institution official alleging foreign interference and data manipulation through X’s algorithms.

This month, the case was escalated to a national police cybercrime unit [pdf], with prosecutors seeking to examine suspected tampering with automated data systems and “fraudulent data extraction.”

In a post, X stated it “categorically denies” the accusations and has declined to turn over requested data.

“French authorities have launched a politically-motivated criminal investigation into X over the alleged manipulation of its algorithm and alleged ‘fraudulent data extraction,'” the company wrote.

“X believes that this investigation is distorting French law in order to serve a political agenda and, ultimately, restrict free speech.”

X claimed that it remains unclear on the exact nature of the allegations but maintained its decision not to comply with the data request, citing its legal right to refuse.

According to a spokesperson for the Paris prosecutor’s office, authorities have only requested access to X’s algorithm – not private user data – to perform a technical verification based on concerns raised by experts and researchers.

The spokesperson told CNBC that investigators are bound by strict confidentiality and that access would be provided through a “secure process” to ensure data protection.

However, X questioned the neutrality of the investigation, accusing the two appointed analysts – David Chavalarias, director of the Paris Complex Systems Institute (ISC-PIF), and Maziyar Panahi, an AI platform leader at the same institution – of being biased.

The company pointed out that Chavalarias runs a public campaign called “Escape X,” urging users to leave the platform, and that both researchers have collaborated on projects that X claims reflect “open hostility” toward it.

Meta pushes back on EU’s AI regulation efforts

Meanwhile, in a separate regulatory standoff in Brussels, Meta has announced it will not sign the European Union’s newly published Code of Practice for general-purpose AI (GPAI) models.

The voluntary code, released on 10th July, is intended to help AI developers comply with the upcoming AI Act – the EU’s comprehensive regulation on AI – which takes effect on 2nd August 2025.

Joel Kaplan, Meta’s global affairs chief, stated on LinkedIn that the company has “carefully reviewed” the EU code but found it rife with “legal uncertainties” and obligations that “go far beyond the scope of the AI Act.”

“Europe is heading down the wrong path on AI,” Kaplan wrote.

While voluntary, the code is designed to offer early signatories reduced regulatory scrutiny.

Signatories are expected to increase transparency around model training, security risks, and copyright compliance – all of which will be legally mandated once the AI Act is enforced.

The EU can impose fines of up to 7% of a company’s global annual turnover for violations under the AI Act, making compliance a high-stakes issue.

Meta’s rejection of the code mirrors broader concerns expressed by over 45 major organisations, which recently urged the EU to delay implementation of the AI Act by two years, citing ambiguity around compliance requirements.

In contrast, Microsoft appears poised to support the EU’s framework.

“I think it’s likely we will sign. We need to read the documents,” Microsoft President Brad Smith told Reuters.

“Our goal is to find a way to be supportive and at the same time one of the things we really welcome is the direct engagement by the AI Office with industry,” he said.

OpenAI and France’s Mistral have already signed the EU code.

The Trump administration has on several occasions criticised Europe over its regulation of US technology companies, equating them with censorship.