Researchers from four universities have identified 27 patterns of corporate influence that major AI companies use to shape legislation and public oversight — tactics strikingly similar to those historically employed by Big Tobacco, Big Pharma and Big Oil.
Major artificial intelligence companies are deploying a familiar playbook to shape the laws meant to govern them, according to a new peer-reviewed study that draws direct comparisons to influence campaigns waged by the tobacco, pharmaceutical and fossil fuel industries.
Researchers from the University of Edinburgh, Trinity College Dublin, TU Delft and Carnegie Mellon University analyzed 100 news articles published around four major AI policy events between 2023 and 2025 — the EU AI Act negotiations and global AI summits held in the UK, South Korea and France. Across that coverage, they documented 249 instances matching 27 established patterns of what scholars call “corporate capture,” a process by which regulatory bodies and public institutions come to serve corporate interests rather than the public good.
The Tactics at Play
Among the most frequently observed methods was “narrative capture” — coordinated efforts to steer how policymakers and the public think about AI regulation. The most common framing was the argument that “regulation stifles innovation” alongside warnings about bureaucratic “red tape.” Researchers say these talking points serve a strategic function: casting oversight as inherently harmful primes audiences for more explicit calls for deregulation down the line.
A second major pattern, which the researchers label “elusion of law,” involves disputed interpretations of existing antitrust, privacy, copyright and labor law — effectively blurring legal lines to the industry’s advantage.
The study also catalogues more direct forms of pressure: lobbying campaigns, retaliation against whistleblowers, researchers and lawmakers who push for stricter rules, and a revolving-door dynamic in which former government policymakers move into advisory or employment roles at leading AI firms. Additionally, the researchers found evidence of significant political donations by AI companies and cases where public officials held equity stakes in the very companies they were responsible for regulating.
Why It Matters
For college students entering a workforce and society increasingly shaped by AI systems, the study raises urgent questions about who actually controls the technology’s trajectory. If the rules governing AI are written under heavy industry influence, the public interest — including issues like data privacy, algorithmic fairness and job displacement — may take a back seat to corporate priorities.
Senior author Zeerak Talat, a Chancellor’s Fellow at the University of Edinburgh’s School of Informatics, pointed out that the parallels to broader democratic concerns were striking.
“It’s remarkable how the findings relate to common experiences of companies having greater influence over democratic processes than people,” Talat said in a news release. “While we cannot draw a causal relationship between attempts at corporate capture and the disenfranchisement of citizens, the former certainly seems to hint at the latter.”
Lead author Abeba Birhane, the director of Trinity College Dublin’s AI Accountability Lab, highlighted which tactics appeared most often in the data.
“In addition to ‘narrative capture’ and the violations and contentious interpretations of antitrust, privacy, copyright and labour laws that were most recurrent, we also found that Big AI frequently uses the notion that ‘regulation stifles innovation’ and that ‘red tape can stymy national interest’ to rationalise their control of the overall narrative,” Birhane said in the news release.
Lessons From Other Industries
The research team argues that history offers a roadmap for pushing back. Drawing on how advocacy communities eventually countered disproportionate industry influence in tobacco, pharmaceutical and oil sectors, the researchers call for clear separation between public and private interests in AI policymaking. They also advocate for binding rules governing how government officials may interact with industry representatives, as a check on conflicts of interest.
The study is scheduled to be presented at the ACM Conference on Fairness, Accountability, and Transparency in June 2026. It is currently available as a preprint on arXiv.
Source: University of Edinburgh