Brussels – One in three responses to the European Commission’s public consultation on stricter tobacco legislation is said to have been generated by artificial intelligence. The allegation – which is worrying because it also casts doubt on the integrity of a democratic instrument such as the EU’s public consultations – comes from two Dutch broadcasters, Pointer KRO-NCRV and NOS, which point the finger at Philip Morris International. According to their investigations, the tobacco multinational is alleged to have acted “in a rather underhand manner” through the ‘Your Voice, Your Choice’ platform, which the company launched to encourage participation in the consultation. In fact, it is alleged to have directed users, via QR codes on yellow posters displayed in tobacco stores and the IQOS website, to a tool offering around 50 pre-defined responses, almost all of which opposed stricter regulations on flavourings, nicotine pouches, filters, and advertising.
When the platform was launched in May of last year, Philip Morris had defined ‘Your Voice. Your Choice’ as a tool to facilitate democratic participation in EU public consultations. “We are offering the ‘Your Voice. Your Choice’ platform, designed to make it easier for individuals to engage in public consultations,” the multinational firm had said in a statement. It had added that the platform “uses an artificial intelligence tool to help users express their opinions, guiding them in structuring and preparing their submissions. Importantly, users remain fully in control: they can edit the output, and what they ultimately write—and whether they choose to submit a response at all – it is entirely up to them. Contributions remain the user’s own at every stage,” according to the company’s statement.
The European Union is working to revise the tobacco legislation by amending the Tobacco Products Directive and the Tobacco Advertising Directive. Brussels is set on tightening the rules on products such as e-cigarettes, nicotine pouches, and heated tobacco. It is standard practice for the Commission, at stages such as these in the drafting of new regulations, to seek the views of citizens and businesses. In this case, over 80,000 responses have been received. Pointer has “submitted over 65,000 responses from fifteen EU countries to the Pangram artificial intelligence detector” to determine whether, and if so which, had been generated by AI. “After Portugal, and just ahead of France, the highest proportion of AI-generated responses comes from us. Of the 786 Dutch responses, 71 per cent were almost certainly written entirely by AI. For all responses from the fifteen EU countries, this percentage stands at 35 per cent,” he explains.
“By way of comparison, in 97 per cent of these types of consultations, the number of responses remains below one thousand. Our research shows that the huge number of responses to the tobacco bill is largely attributable to an artificial intelligence model; the vast majority of Dutch responses appear to be linked to an artificial intelligence campaign by Philip Morris (PMI),” the Pointer article noted. According to the National Committee Against Smoking, France’s oldest association for the prevention of smoking, the high number of responses to the EU Commission’s public consultation “may have been perceived as a sign of unprecedented civic mobilisation, but in reality it masks a manipulation campaign orchestrated by the tobacco industry.”
Philip Morris International’s response was not long in coming. “We reaffirm our commitment to informing and facilitating the participation of EU citizens and stakeholders in the democratic process, as well as to using technology to support this objective. Users of the Voice tool retained full control over their contributions at all times: they were free to edit the text and decide what to include or exclude,” the company said. “Anti-tobacco lobbyists are attempting to discredit thousands of citizens and stakeholders who chose to take part in the consultation on the Tobacco Products Directive (TPD) – with or without the support of artificial intelligence – by spreading fake news,” it added. According to the company, “their aim is clear: to delegitimise viewpoints that challenge their preferred narrative, diverting attention away from the outcome of the consultation and focusing it on a contrived controversy.” Furthermore, “what is at stake is the right of stakeholders and citizens to participate in the democratic decision-making process without being ignored or silenced because their views are inconvenient for a particular ideological agenda.” In this context, “PMI remains committed to transparency, to the participation of people and stakeholders, and to ensuring that individuals have the tools and opportunities to engage with EU policymakers and make their voices heard”, it concludes.
![[Foto: Unsplash]](https://www.eunews.it/wp-content/uploads/2026/06/pawel-czerwinski-7USMFYqt1NI-unsplash-350x250.jpg)
