MCA and Meta partner to launch WhatsApp helpline against AI-generated misinformation

The Misinformation Combat Alliance (MCA) and Meta are working on launching a fact-checking helpline on WhatsApp to combat media generated using artificial intelligence which may deceive people on matters of public importance, commonly known as deepfakes, and help people connect with verified and credible information. The helpline will be available for the public to use in March 2024.

The industry leading initiative will allow MCA and its associated network of independent fact-checkers and research organisations to address viral misinformation- particularly deepfakes. People will be able to flag deepfakes by sending it to the WhatsApp chatbot which will offer multilingual support in English and three regional languages (Hindi, Tamil, Telugu).

The MCA will set up a central ‘deepfake analysis unit’ to manage all inbound messages they receive on the WhatsApp helpline. They will work closely with member fact-checking organizations as well as industry partners and digital labs to assess and verify the content and respond to the messages accordingly, debunking false claims and misinformation.

The focus of the program is to implement a four-pillar approach – detection, prevention, reporting and driving awareness around the escalating spread of deepfakes along with building a critical instrument that allows citizens to access reliable information to fight the spread of such misinformation.

Commenting on the partnership, Shivnath Thukral, director, Public Policy India, Meta, “We recognise the concerns around AI-generated misinformation and believe combatting this requires concrete and cooperative measures across the industry. Our collaboration with MCA to launch a WhatsApp helpline dedicated to debunking deepfakes that can materially deceive people is consistent with our pledge under the Tech Accord to Combat Deceptive Use of AI in 2024 Elections. As a company that has been at the cutting edge of AI development for more than a decade, we remain committed to work with industry stakeholders to introduce common technical standards for AI detection, transparency solutions and policies, along with empowering people on our platforms with resources and tools that make it simpler for them to identify content that has been generated using AI tools and curb the spread of misinformation.”

“The Deepfakes Analysis Unit (DAU) will serve as a critical and timely intervention to arrest the spread of AI-enabled disinformation among social media and internet users in India. Its formation highlights the collaboration and whole-of-society approach to foster a healthy information ecosystem that the MCA was set up for. The initiative will see IFCN signatory fact-checkers, journalists, civic tech professionals, research labs and forensic experts come together, with Meta's support. We hope the DAU will become a trusted resource for the public to discern between real and AI generated media and we invite more stakeholders to be a part of the initiative,” said Bharat Gupta, president, Misinformation Combat Alliance.

Meta’s fact-checking program in India includes partnerships with 11 independent fact-checking organisations that help users to identify, review, verify information and help prevent the spread of misinformation on its platforms. On WhatsApp, users can double-check information that sounds suspicious or inaccurate by sending it to WhatsApp tiplines. People can also follow fact-checking organisations on WhatsApp Channels to receive verified, accurate and timely updates. In addition to the fact-checking program, WhatsApp addresses misinformation by limiting forwards and actively constraining virality on the platform.

The approach to addressing deceptive synthetic media at Meta has several components, including working to investigate deceptive behaviors like fake accounts and misleading manipulated media; its third-party fact-checking program, in which fact checkers rate misinformation, including content that has been edited or synthesised in a way that could mislead people; and engaging with academia, government and industry. Meta has recently announced an AI labeling policy. In the coming months, it will label images that users post to Facebook, Instagram and Threads when it can detect industry standard indicators that they are AI-generated.