Iran rolls out AI platform prototype amid facial recognition surveillance accusations

Iran has presented a prototype of its national AI platform, designed to address both the country’s lagging technological development and the threat of international sanctions.

The AI platform will see a phased rollout with the final version expected to be launched in March 2026, Hossein Afshin, the Iranian Vice-President for Science, Technology and Knowledge-Based Economy, said during its launch in Tehran on Saturday.

The platform is based on open-source technology and domestic infrastructure which ensures its functionality even during internet disruptions and sanctions, according to officials.

“We are moving forward with the world,” says Hamidreza Rabiei, head of the Advanced Information and Communication Technology Research Institute (AICTC). “We are not taking any API from any foreign platform, and if the internet is cut off, nothing will happen to the platform because we are connected to the national internet.”

The project kicked off in October last year and involves several institutions, including the Sharif University of Technology, which is sanctioned by several governments. The platform includes a GPU-based processing infrastructure, large language and multi-modal models, intelligent agents and application layers for various industries, according to media outlet Iran International.

UN details Iran’s use of AI to crack down on protest

The presentation of the initiative comes days after the United Nations Human Rights Council published its latest report on Iran’s rights violations during recent anti-hijab protests, which includes using facial recognition surveillance against women refusing to wear hijabs.

The protests against the Islamic country’s strict hijab rules kicked off following the death of a young woman named Mahsa Amini in police custody in September of 2022. The demonstrations were followed by a violent crackdown by police forces – some of which amounted to crimes against humanity, according to the UN’s fact-finding mission.

In April 2024, the state monitored hijab compliance using aerial drone surveillance, while the government also reportedly installed facial recognition software at the entrance of the Amirkabir University in Tehran to monitor female students.

The UN mission also analyzed the Nazer mobile application which allows police forces and vetted individuals to take photos of women flouting hijab rules. The government-made app allows users to add the location, date, time and license plate connected to the alleged hijab infraction. The app then flags the vehicle to the police and triggers a warning text message to the registered vehicle owner, according to the UN’s report.

The Iranian regime has also promised to cross-check social media posts from women without hijabs with the national ID database and allocated funds to install more cameras.

How Iran acquires its surveillance technology

Experts are still uncertain about the abilities of Iran’s law enforcement to apply facial recognition, noting that the technology is being used to scare people into compliance. Despite this, the government has been evolving in its use of AI-based surveillance technologies, according to a 2024 report released by the European Parliament.

Cameras with facial recognition have been strategically positioned to capture public activity. The deployments have been recorded during periods of civil unrest such as the Iranian protests of 2019-2020 and the recent anti-hijab protests.

Iranian authorities have procured the surveillance technology through partnerships with China, Turkey, India and the United Arab Emirates (UAE).

The largest exporter of components, including most of the hardware infrastructure, is the UAE, the report notes. Media investigations have also noted the involvement of Chinese surveillance providers such as Tiandy Technologies and Dahua. In 2022, Tiandy joined Dahua and Hikvision on the U.S. Commerce Department blacklist for allowing the sales of U.S.-made technology to Iran’s Revolutionary Guards.

Iran has also been relying on homegrown facial recognition suppliers, according to the U.S. national security think tank Center for Intelligence Research and Analysis (CIRA).

This includes facial recognition firms Rayvarz, Sciences Green, which provides services to national and local government entities, and UID which has developed products for law enforcement agencies. With the support of the Iran Innovation and Prosperity Fund, the company created authentication hardware Maskinow which is used to identify masked and unmasked individuals.

Last year, a hacktivist group called GhostSec claimed to have breached a facial recognition software system that Iranian authorities were allegedly using for mass surveillance.

Article Topics

biometric identification  |  biometrics  |  EU  |  facial recognition  |  Iran  |  mobile app  |  Nazer app  |  United Nations  |  video surveillance