Why Data Should Be Shared

Source details

Economic innovation in the AI age could stall without regulations that mandate access to data.
AI amplifies the “feedback effect”: more data → better products → more customers → more data, creating self-reinforcing market concentration.
Data access mandates (especially for nonpersonal data), if cleverly designed, can reduce transaction costs and enable broader use of data, rewarding ideation rather than data hoarding.

Google trained its spellchecker some 20 years ago using typos from billions of users; no competitor could match that access to data.
Google accounts for 9 in every 10 internet searches.
Six Big Tech companies—Alphabet (Google), Netflix, Meta, Apple, Amazon, and Microsoft—account for almost half the world’s internet traffic.
Four companies—Alphabet, Microsoft, Meta, and Amazon—dominate AI computing capacity.
Surveys show that at least 80 percent of what’s collected worldwide is never used.
Economic theory and empirical work: economies of scale and scope favor larger firms; dominant firms may lose interest in innovation and prioritize protecting market share (citing work such as Ufuk Akcigit and commentary by Cory Doctorow).

Antitrust and competition regulation that break up platforms address symptoms but not the underlying dynamic that rewards those with the most data.
Policies granting individuals more control over personal data (e.g., GDPR-style approaches) routinely fail to counter concentration because of collective-action problems—many care but few exercise control.
Legal ownership or exclusion rights over data and complex licensing regimes risk reducing overall data accessibility and disproportionately burdening individuals and small start-ups.

Mandates can eliminate the need for negotiated licenses, lowering transaction costs and enabling smaller firms to access data.
Access mandates encourage data holders to extract value by using the data themselves, promoting data-savviness among conventional firms.
Multiple players applying ideas to the same data increases ideation and innovation rather than rewarding mere possession.
Historical and existing precedents:
- GPS (US military) and EU Galileo: public locational data availability produced safety, efficiency, sustainability gains and enabled a multibillion-dollar industry.
- Legal requirements to publish financial results and emissions data are longstanding examples of mandated data disclosure.
- EU rules now require large digital platforms to share some data with smaller competitors.
- US antitrust settlements have required companies to let competitors access their data; Google was recently required to do so as part of an antitrust trial.
- 1950s antitrust settlement required AT&T to let US firms use its transistor patents for free, enabling start-ups to design integrated circuits and helping bootstrap Silicon Valley.
- The patent system itself grants temporary exclusivity only if inventors share details, illustrating value from mandated disclosure.

Prioritize regulations that mandate access to data (with emphasis on nonpersonal data) as a promising intervention to reduce concentration and sustain innovation.
Design access mandates to:
- Reduce transaction costs by avoiding complex licensing negotiations.
- Enable smaller firms and start-ups to compete and innovate using shared data resources.
- Promote use of data over mere accumulation, encouraging conventional firms to develop data capabilities.
Recognize trade-offs: ownership- and license-based regimes can restrict access and entrench larger firms; antitrust alone may be insufficient without addressing data-access dynamics.

Viktor Mayer‑Schonberger, Professor of Internet Governance and Regulation at Oxford University’s Internet Institute; F&D Magazine, December 2025.

Why Data Should Be Shared
- Why Data Should Be Shared (Markdown version)
- Why Data Should Be Shared (PDF)