Why IBM Is Pressuring Governments to Restrict Facial Recognition Exports

IBM (International Business Machines Corporation) recently sent a letter to the U.S. Department of Commerce urging the restriction of facial‑recognition system exports, especially to countries that might use them for large‑scale surveillance, racial discrimination, and human‑rights violations.

IBM distinguishes between one‑to‑one facial‑recognition, such as phone unlocks or electronic boarding passes, which it does not oppose, and one‑to‑many systems that match a single image against a database of faces, which it seeks to limit.

According to IBM’s Vice President of Government and Legal Affairs Christopher Padilla, one‑to‑many systems can identify individuals in crowds without consent, and IBM stresses that data training should only occur with explicit permission.

IBM argues that controlling online access to training data is an effective way to prevent large‑scale monitoring and the extraction of raw data for mass surveillance.

The company also recommends referencing multilateral agreements such as the Wassenaar Arrangement to strengthen global cooperation and prevent the use of facial‑recognition technology for regime repression.

Several U.S. cities—including San Francisco, Oakland, San Diego, and Portland—have already banned facial‑recognition use, citing its limitations, lack of standards, and potential racial bias.

Reuters reports that the European Union is moving to prohibit facial‑recognition in public spaces such as parks, tourist sites, and sports venues for at least five years, allowing time for regulated legislation to prevent misuse.