India platform regulation

Meta's India-Specific Instagram Tuning Is a Sensible Fix, but the Pressure Behind It Runs Through Informal Advisories

Meta says it tuned Instagram recommendations for India to cut sexual and violent content. The safety goal is fair, but MeitY's advisory route lacks a clear legal basis.

Meta, MeitY and Instagram Recommendations People of Internet Research · India ~50% fewer Users reporting recommended content Meta says people report recommende… 7 Days to answer MeitY notice MeitY gave Meta seven days to expl… 4 Cases of restriction cited IFF lists four accounts or posts r… peopleofinternet.com
Meta, MeitY and Instagram Recommendati… People of Internet Research · India ~50% fewer Users reporting recommended cont… 7 Days to answer MeitY notice 4 Cases of restriction cited peopleofinternet.com

Key Takeaways

On 16 November 2026, Meta published a post titled "An Update on the Content We Recommend on Instagram in India". It says the company has "tuned" its systems "specifically for India" so they keep more "sexually suggestive content and content that depicts violence" out of recommendations. MediaNama reported the change on 18 November and noted that Meta's post does not mention any Indian government ministry. It also noted that the announcement arrives amid heavy official scrutiny.

The change is defensible on its merits. The sequence of pressure around it is the part that deserves scrutiny.

The strongest case for the government's concern

The case for intervention is serious. In early July 2026, a BBC investigation reported Instagram ads using terms such as "rape video" and "child video" that led to Telegram channels selling child sexual abuse material. It also reported sexually suggestive recommendations surfacing for users who had not searched for them. According to The Week's report of 5 July, MeitY issued a notice demanding that Meta disable all ads promoting or facilitating access to such material. It gave Meta seven days to submit a detailed explanation.

A state that acts on child sexual abuse material (CSAM) in ads is doing what states should do. Recommendation systems are also a legitimate regulatory target. They are product design choices made by the platform, not passive hosting of user speech. If an algorithm pushes sexualised or violent material at users who did not ask for it, a regulator has a fair interest in how that system behaves.

What Meta actually announced

Meta's post describes two changes. The first is better detection, using "more capable AI models", which Meta says applies worldwide. The second is the India-specific tuning, which it says reflects what "people here told us they didn't want to see". The only figure in the post is Meta's claim that people now report recommended content "about half as often" as a year ago. That figure is self-reported and unaudited, and the post gives no baseline or methodology.

The post also points to existing user controls: Sensitive Content Control, Political Content Control, Not Interested and Hidden Words. It notes that teen accounts sit in a 13+ content setting. Those are the proportionate tools of the open internet. They put adjustable defaults in users' hands, and they are what the evidence favours over blunt takedown mandates.

Why the route matters

Two separate government actions sit behind this story, and they should not be conflated.

The July action concerned CSAM in ads. It is about plainly illegal material and has a clear public-interest basis.

The October action was different. MeitY's Cyber Laws Division issued an advisory to Meta dated 10 October, according to the Internet Freedom Foundation. The press reported it on 18 and 19 October. As All India Radio summarised it, the government asked Meta to ensure its recommendation systems do not cause "disproportionate exposure of users to a single or limited set of narratives". It also urged stepped-up checks on content that could cause public disorder.

That advisory is about protest-related content and public order. It is not about sexual content, and MediaNama itself says Meta's post does not link the India tuning to any ministry. We cannot say the tuning was a response to the advisory, and we should not claim it was. What we can say is that Meta is operating under overlapping informal pressure and has an obvious incentive to be seen complying.

That is where the legal problem sits. IFF argues that the advisory "is not a legal order" and that MeitY is not empowered to issue such advisories under any statutory provision. IFF is an advocacy group and its account is one-sided, but the underlying principle has solid judicial support. In Shreya Singhal v. Union of India (2015), the Supreme Court read down Section 79(3)(b) of the IT Act. It held that an intermediary's duty to remove content arises on actual knowledge from a court order or a government notification. It also held that any such order must "strictly conform to the subject matters laid down in Article 19(2)".

An advisory that asks a platform to rebalance "narratives" is not obviously tied to any Article 19(2) ground. It also creates no formal process for review or appeal.

The proportionate path

The risks are asymmetric. A rule-bound CSAM or illegal-ads regime gives platforms clear duties and gives users and publishers a way to challenge errors. Informal advisories invite over-compliance. Platforms facing the loss of safe-harbour protection tend to suppress borderline lawful speech first, because that is the cheapest way to reduce risk.

The IFF account lists four cases between 10 and 13 October in which accounts or posts were restricted. These include a news outlet's reel. The cases are unverified beyond IFF's account, but they illustrate the failure mode. Similarly, "violent content" is a category that sweeps in news footage of protests, conflict and police action. An India-specific filter that is tuned to government sensitivities rather than to user preferences could quietly demote journalism.

Three safeguards would keep the child-safety gains and limit the speech costs:

Bottom line

Meta's tuning is a reasonable product decision and likely a net positive for teen safety. The concern is the governance pattern around it, in which legal duties are shaped by summons, notices and advisories rather than by law. Indian users are better served by clear statutory duties on recommendation systems, narrowly aimed at unlawful content, than by a standing threat that platforms will adjust their algorithms to whatever the ministry finds uncomfortable.

Sources & Citations

  1. Meta: An Update on the Content We Recommend on Instagram in India
  2. Shreya Singhal v. Union of India (Supreme Court, 2015)
  3. All India Radio: Govt asks Meta to prevent disproportionate exposure to limited narratives
  4. MediaNama: Meta tuned algorithms for India
  5. The Week: Centre orders Meta to remove Instagram ads promoting CSAM
  6. Internet Freedom Foundation: MeitY's advisory to Meta must be withdrawn