Meta has begun acting on the Centre’s concerns over CSAM, deepfakes and unlabelled AI content after multiple rounds of government discussions.
The move signals tighter government scrutiny of social media platforms and greater accountability for content moderation in India.
The Centre has pushed Meta to strengthen CSAM detection, human moderation and AI-content labelling, while India’s amended IT rules impose stricter takedown and grievance-redressal timelines.
Social media giant Meta has reportedly begun taking steps to address concerns raised by the Central government over child sexual abuse material (CSAM), deepfakes, unlabelled AI-generated content, among others, hosted on its platform following several rounds of discussions with officials.
As per a report by PTI, this comes after discussions between the social media major and the Central government over the former’s handling of CSAM, with the government pressing the company to strengthen its identification and removal of unlawful material. Citing sources, the report added that Centre has told Meta that CSAM violates Indian law and that the company could lose safe-harbour protection if it fails to meet its legal obligations.
Meta’s global team has also explained how its algorithms and other systems operate and outlined measures it plans to introduce in response to the government’s concerns. Some of the problems, officials acknowledged, require “judgment calls and tweaks”.
As per a report by PTI, the Centre has sought to draw a distinction between regulation and censorship. As per government sources, its aim is to ensure that platforms comply with Indian law and limit the harms arising from social media, rather than restrict legitimate coverage.
“The message has gone home,” the sources were quoted as saying by PTI.
One unresolved issue is the persistence of harmful AI-generated material. The government has questioned why such content can reappear after being flagged, and why AI-generated videos without appropriate labels remain accessible on Meta’s platforms.
In an earlier meeting, officials also pressed Meta to increase human oversight of content moderation, particularly for Indian languages and material shaped by local cultural and social contexts.
The government is separately examining WhatsApp’s proposed username feature. Officials said the issue is broader than WhatsApp and could have implications for other messaging services.
The latest development marks another stage in the Centre’s scrutiny of social media intermediaries. Earlier this month, the company’s global team landed in India after MeitY summoned its global public-policy leadership over Facebook’s temporary removal of a video posted by Prime Minister Narendra Modi.
The video, posted on July 23 amid student protests over alleged paper leaks and irregularities in NEET examinations, was PM Modi’s first address to students during the protests. Facebook restored it several hours later, attributing the removal to an error in its content-moderation systems.
The episode triggered a series of discussions between Meta and the Centre, eventually leading to an apology from the company. While Meta’s chief global affairs officer Joel Kaplan met with the Centre over the issue, government sources told PTI that CEO Mark Zuckerberg had also apologised over the broader content-moderation lapses, including CSAM and deepfakes
Meanwhile, Elon Musk-owned X will have to comply with Indian law even as it rolls out a new policy to make government requests for content takedowns public, a government official said, as per PTI report.
This comes after Musk said X had introduced a new algorithm under which “any censorship required by the governments is now clearly visible”. The Centre, however, has made clear that the policy will not override India’s legal requirements.