Meta Reduces Violent and Sexually Suggestive Instagram Recommendations in India

9 Min Read
Meta Reduces Violent and Sexually Suggestive Instagram Recommendations in India

Instagram changes what it puts in front of Indian users

Meta said on November 16, 2026, that it had strengthened Instagram's recommendation systems in India to show less sexually suggestive content and material depicting violence. The company says users now report recommended posts about half as often as they did a year earlier, although it has not published the underlying figures.

Contents
  1. Instagram changes what it puts in front of Indian users
  2. What changes across Reels, Explore and Feed?
  3. How the AI checks work
  4. What the government asked Meta to review
  5. The protests behind the regulatory scrutiny
  6. Recommendations and removal follow different rules
  7. What does the claimed drop in reports tell users?
  8. Which settings can users change?
  9. Teen Accounts add another layer of protection
  10. Key Points

The changes affect posts suggested from accounts people do not follow, including material appearing in Reels, Explore and Feed. More capable artificial intelligence models now assess content before Instagram recommends it, rather than waiting for someone to report or hide it.

Meta described the changes in its India recommendations update as work carried out over the past year in response to user feedback. The announcement comes after India's Ministry of Electronics and Information Technology (MeitY) asked the company to review how its platforms rank and promote posts, particularly around public events and situations involving public order.

For users, the practical change concerns discovery: which unfamiliar creators, videos and images Instagram puts in their path. For creators, it concerns how far a post can travel beyond an existing audience. Posts that break Instagram's Community Standards continue to be removed, while the recommendation rules determine which other posts receive wider exposure.

What changes across Reels, Explore and Feed?

Instagram recommends content by using signals about a person's interests and activity. Meta gives a simple example: someone interacting with restaurants and bookstores may receive suggestions about food, recipes, books and reading.

Reels, Explore and parts of Feed introduce users to accounts they have not chosen to follow. Instagram applies its Recommendations Guidelines to this material, seeking to keep content it considers objectionable, particularly sensitive or of low quality out of those suggestions.

Meta, the company that owns Instagram and Facebook, described the changes tailored to India in its announcement:

We've improved our systems specifically for India so they catch more of what people here told us they didn't want to see.

The company said those systems now keep more sexually suggestive material and depictions of violence out of recommendations. Users may consequently encounter fewer such posts while scrolling through suggested content.

Advertisement

How the AI checks work

Meta says its systems around the world check posts against the Recommendations Guidelines before recommending them. The latest improvements include more capable AI models that assess content directly, allowing Instagram to act before a user complains.

An AI model is software trained to identify patterns. Here, its job is to help assess whether a post meets Instagram's rules for recommendations. Meta says it has also adjusted its systems specifically for the kinds of content people in India have said they do not want to see.

User feedback remains part of the process. Hiding a post or selecting "Not Interested" tells Instagram something about an individual's preferences and helps the company identify material that audiences find unwanted.

The approach combines a default filter with personal choices. Instagram screens content before suggesting it, while a user's actions can further shape what appears in future recommendations.

What the government asked Meta to review

MeitY's advisory was dated October 10, according to accounts describing its contents; another account dates an advisory seeking greater diversity in recommendations to October 19. The ministry asked Meta to review its ranking and recommendation systems so misleading, manipulated or out of context material would not receive disproportionate visibility.

The government's concerns went beyond sexual or violent imagery. It also wanted stronger checks on potentially inflammatory posts, adequate context for photographs and videos whose authenticity or circumstances were unclear, and information about the steps Meta had taken.

MeitY also raised concerns about users being shown a single or limited set of narratives, especially around public events and situations involving law and order. In simple terms, the concern was that repeated recommendations could keep putting similar versions of an event in front of a person while other relevant information received less attention.

Meta has not directly linked the Instagram changes to the advisory. Its explanation is that the improvements developed over the preceding year through user feedback. The ministry's request covers the distribution of public information, while Meta's announcement concentrates on unwanted recommendations, including sexual and violent material.

Advertisement

The protests behind the regulatory scrutiny

Supporters of the Cockroach Janta Party (CJP) began protests in Mumbai on October 2, demanding the resignation of Chief Election Commissioner Gyanesh Kumar. The group accused the Election Commission of undermining voter rights through the Special Intensive Revision of electoral rolls, a process for reviewing voter lists.

The protests later spread to other cities and generated videos, memes and other social media coverage. The October 10 advisory did not name CJP or a particular incident.

Government scrutiny of Meta had already included a dispute over moderation. In July 2026, MeitY summoned Joel Kaplan, Meta's president of global affairs, over the company's brief blocking of a reel posted by Prime Minister Narendra Modi. The ministry also sought an explanation that month over advertisements promoting child sexual abuse material, with Meta and Google subsequently saying they would report related incidents to Indian authorities.

The main dates show how the current announcement fits into that sequence:

  • July 2026: MeitY challenged Meta over moderation and advertisements promoting child sexual abuse material.
  • October 2: CJP supporters began the latest protests in Mumbai.
  • October 10: MeitY's advisory, as dated in most accounts, asked Meta to review ranking and recommendations.
  • November 16: Meta published its update on Instagram recommendations in India.

Recommendations and removal follow different rules

Instagram's Community Standards govern what users are allowed to post. Meta says material violating those standards is removed, while content unlawful in a particular country is restricted there.

The Recommendations Guidelines govern another decision: whether Instagram should actively introduce a post to people who do not follow its creator. Content can remain accessible while being excluded from recommendations because Instagram considers it particularly sensitive or otherwise unsuitable for wider promotion.

That gives the change a concrete effect for creators. A post may remain on an account while receiving less exposure through Explore, Reels or Feed suggestions. For viewers, fewer recommendations mean fewer opportunities to encounter that material while browsing unfamiliar accounts.

Meta has previously changed Facebook's ranking of political content in countries including Canada and Brazil. The India announcement concerns Instagram's recommendation filters, alongside separate controls that let users reduce suggested political posts.

Advertisement

What does the claimed drop in reports tell users?

Meta says people in India now report recommended content about half as often as they did a year earlier. The company attributes that decline to stronger protections and its ability to keep more unsuitable posts out of recommendations before users encounter them.

The measure describes how frequently users report recommendations. Meta has not disclosed the starting or ending figures, its calculation method, the volume of content affected, the share of recommendations filtered or a schedule for further changes.

Those missing details leave readers without a numerical picture of how much material Instagram screened out. The company's published claim is a reduction in reporting frequency, rather than a count of posts removed or recommendations prevented.

Meta says it publishes policy enforcement information through its global Community Standards Enforcement Report and monthly transparency reports for India under the IT Rules, 2021. These are the reporting channels it identified for information about enforcement.

Which settings can users change?

Instagram offers several controls alongside its default recommendation filters. Sensitive Content Control is available in Settings and Activity, under Content Preferences. It lets people choose to see less sensitive material across Search, Reels, Explore, suggested accounts and recommendations in Feed.

Meta says the most mature content is already excluded from recommendations in India. Sensitive Content Control adjusts what users see within that limit. Political Content Control, also under Content Preferences, lets people reduce political recommendations from accounts they do not follow.

The remaining tools address different parts of the experience:

  • Not Interested: Available through the three dots on a post or reel, this sends feedback that a recommendation is unwanted.
  • Hidden Words: Filters selected words, phrases and emojis from comments and message requests.
  • Block: Prevents an account from finding a profile, seeing posts or sending messages, and covers new accounts that person tries to create.

Users can choose a control according to the problem they face. Sensitive Content Control changes the kinds of suggestions they receive; Hidden Words filters incoming text; blocking limits contact from particular people.

Advertisement

Teen Accounts add another layer of protection

Meta says Teen Accounts, introduced in India in 2025, automatically place teenagers in a 13+ content setting shaped by the company's guidelines for age appropriate content.

Teenagers therefore receive additional content settings alongside the recommendation protections applied more broadly. The 13+ setting is a content category within Instagram's teen protections, intended to shape what younger users encounter.

Meta says it will continue adjusting its systems as technology and user behaviour change. The immediate changes for Indian users are stronger screening before recommendations appear, supported by controls for personal preferences and additional settings for teenagers.

Key Points

  • Meta's November 16 update says Instagram is recommending less sexually suggestive and violent content in India.
  • The changes affect suggestions from unfamiliar accounts across Reels, Explore and Feed.
  • More capable AI models assess content before users report or hide it.
  • MeitY asked Meta to review how its systems distribute potentially misleading and inflammatory material.
  • Meta attributes the changes to user feedback and has not directly linked them to the advisory.
  • The company claims reporting frequency has roughly halved, without publishing the underlying figures.
  • Users retain content controls, while Teen Accounts automatically apply a 13+ content setting.
Share This Article

You May also Like