AI Built Into Apps: The Chatbot in Search, Snap, WhatsApp, Games

A parent blocks ChatGPT on a 12-year-old’s phone and considers the job done. The next afternoon, the kid types a history question into Google and gets a full, ready-to-submit paragraph from AI Mode. Then she asks Snapchat’s My AI about a friend fight, and tags @MetaAI in a WhatsApp group to settle an argument.

None of that required opening a “chatbot.” AI built into apps is now the main way children meet generative AI, and the controls are scattered: one app has a single off switch, another has none. This article maps where the assistants live in the apps kids use most, what each one sees, and what you can actually change.

Key Takeaways

• Generative AI now sits inside search, messaging, social and game apps, so blocking chatbot apps leaves most exposure untouched.
• Snapchat’s My AI is one of the few embedded assistants a parent can fully disable, through Family Center.
• Google’s AI Overview and AI Mode cannot be turned off by parents, schools or students, according to Common Sense Media’s July 2026 assessment.
• WhatsApp says Meta AI cannot see personal messages, but receives whatever you or others send to it, including in groups.
• A European Commission proposal reported in September 2026 would require chatbots built into social platforms to be off by default for minors.

Why AI Built Into Apps Is Where Kids Actually Meet It

AI built into apps means a generative assistant that lives inside a product whose main job is something else: searching, messaging, posting, playing. The kid doesn’t download it. It shows up as a search summary, a pinned contact, a button in a chat, or a talking character.

That matters for two reasons. First, exposure starts younger than parents assume. Common Sense Media’s 2026 census of 1,211 young people aged 9 to 17, released October 27, found that 65% had used AI in their free time, including 36% of 9-year-olds who had used a chatbot. Second, kids don’t name it. Ask “Do you use AI?” and a child who reads Google’s AI summaries every day may say no.

A Map of Where the Assistant Lives

Search: Google AI Overview and AI Mode

AI Overview is the generated summary that appears above Google results. AI Mode is a full chat-style search. Both generate answers by drawing on web pages.

Common Sense’s Youth AI Safety Institute rated both “Unacceptable Risk” in a July 14, 2026 assessment. AI Mode gave a complete, submittable answer to all 180 academic assignments tested. When a hotline or medical referral was warranted, AI Overview provided one 58% of the time and AI Mode 77%. Testers found no meaningful difference between 11-year-old and 15-year-old accounts. And the key finding for parents: neither feature can be disabled by parents, schools or students. AI Overview appears on school-issued Chromebooks.

Our explainer on AI search versus traditional search covers how to teach kids to scroll past the summary to the sources.

Snapchat: My AI

My AI is a chatbot pinned inside Snapchat’s chat list. Common Sense rated it “High risk” in August 2024. It is also one of the easiest embedded assistants to control. Snapchat’s support page gives three steps: open Family Center, scroll to Settings, and turn on “Disable My AI.” Snapchat says that once disabled, “My AI will not store or process your teen’s queries, nor reply to your teen’s messages.” That requires a linked Family Center account, which your teen has to accept.

WhatsApp, Instagram and Messenger: Meta AI

Meta AI is Meta’s assistant across its apps. In WhatsApp, it appears as a chat you can open and can be summoned in group chats. Common Sense rated Meta AI “Unacceptable Risk” in August 2025, per its risk assessment list.

WhatsApp’s privacy questions page makes three claims worth knowing. Personal messages “are end-to-end encrypted and are not seen by Meta AI.” Meta AI is “only ‘on’ when you choose to message it,” and only you or a group participant can start it. But “if you choose to chat with Meta AI, directly or in your groups, then Meta receives what you or others send it,” and users “should only send to Meta AI what you’re comfortable with it remembering.”

Read that last line twice. In a group chat, a classmate can send Meta AI something your child wrote. Encryption protects the messages between people, not what anyone chooses to hand to the assistant. These are WhatsApp’s own descriptions; regulators are checking. Brazil’s data authority, the ANPD, opened two monitoring proceedings on August 21, 2026 covering Instagram, WhatsApp, Facebook and Meta AI, SBT News reported.

X: Grok

Grok is xAI’s assistant, built into X and reachable by tagging @grok in replies. Common Sense rated “Grok and @grok on X” Unacceptable Risk in January 2026. A teen with an X account can reach Grok without installing a separate app. Our guide to Grok for parents covers its settings.

Games: talking characters

Fortnite and Roblox now give creators tools to build AI characters that answer players in voice or text, with rules against romantic and mental-health personas. We covered how they work and what is stored in AI characters inside Fortnite and Roblox.

School: Gemini and classroom tools

Many districts have switched on Google’s Gemini for students. Common Sense rated Gemini for K-12 “High risk” in January 2026, in part because a sixth-grader and a senior get the same responses. Here the controls belong to the school’s administrator, not the family, so the question goes to the school.

The Embedded-AI Map at a Glance

App	Where the AI shows up	Common Sense rating (date)	Can a parent switch it off?	What it receives
Google Search	AI Overview summary; AI Mode chat	Unacceptable Risk (Jul 2026)	No, per Common Sense	Search queries
Snapchat	My AI pinned in chat	High risk (Aug 2024)	Yes, “Disable My AI” in Family Center	Teen’s messages to My AI; none once disabled, per Snap
WhatsApp / Instagram / Messenger	Meta AI chat; @Meta AI in groups	Unacceptable Risk (Aug 2025)	Optional to use; WhatsApp says it’s only on when messaged	Whatever you or group members send it
X	Grok, @grok in replies	Unacceptable Risk (Jan 2026)	Not a separate app; account-level	Prompts and tagged posts
Fortnite / Roblox	AI-powered characters	Not rated	Through game maturity and chat settings	Player speech or text, per each platform’s policy
School Chromebooks	Gemini; AI Overview in search	Gemini K-12: High risk (Jan 2026)	School admin decides	Student prompts under school account

What Is Changing in the Rules

The clearest proposal targets exactly this embedded design. El Español reported on September 17, 2026 that the European Commission’s proposed EU Kids Act would require a chatbot built into a platform such as X to be “deactivated by default” for minors, and bar platforms from encouraging its use. Age checks would happen at account creation and could not rely on users simply typing an age; they would have to avoid tracking, for example through zero-knowledge proofs. It is a proposal, its final form is not settled, and outlets differ on whether “off by default” covers all chatbots or only embedded ones. It applies in the EU, not the US.

In the US, the closest thing is state law. New York’s chatbot law, signed October 28, 2026, bars features such as pretending to be human or using a child’s personal information from earlier conversations for users under 18, but it does not switch embedded assistants off.

What to Do at Home

Run a ten-minute app audit with your kid

Open each app on your child’s phone together and look for the assistant: the sparkle icon, the pinned bot, the “Ask Meta AI” bar. Ask your child to show you where they’ve used it. You’ll learn more from their tour than from any settings page, and they’ll learn you’re curious rather than hunting.

Switch off what can be switched off

Snapchat’s My AI is the clearest case. For apps with no switch, the setting is behavioral: agree that Google’s summary is a starting point, and that homework answers come from the kid.

Set the group-chat rule

Tell your child: anything in a group can be sent to an AI by anyone in the group. Don’t put things in group chats you wouldn’t want an assistant to remember.

Ask the school one question

“Which AI tools are switched on for my child’s school account, and what can they do?” The district controls that layer.

What not to do

Don’t assume blocking a chatbot app is enough, and don’t try to block Google entirely. You’ll lose a useful tool and push the search elsewhere. Teaching your kid to recognize an AI answer, and check it, travels to every app.

What to Watch For Over the Next 3 Months

• Week 4: You and your child have a written list of every app on their phone that contains an AI assistant, and which ones are disabled.
• Month 2 red flags: Homework that reads like a polished summary, a new “friend” contact that turns out to be a bot, or your child asking an assistant about personal problems in a group chat.
• Month 3 self-check: Check whether the EU proposal or any US state has moved embedded assistants to “off by default” for minors, and whether Google has added a parent control for AI Mode.

Frequently Asked Questions

Can I turn off Google AI Overviews for my child?

Not with a parental setting, according to Common Sense Media’s July 2026 assessment, which found neither AI Overview nor AI Mode can be disabled by parents, schools or students. What you can do is teach your child to treat the summary as a lead and open the sources underneath.

How do I turn off My AI on Snapchat?

Through Family Center: open it, scroll to Settings, and turn on “Disable My AI.” Snapchat says My AI then won’t store or process your teen’s queries or reply to them. Family Center requires your teen to accept the link with your account.

Can Meta AI read my child’s WhatsApp chats?

WhatsApp says personal messages are end-to-end encrypted and not seen by Meta AI. But Meta receives anything sent to Meta AI, including in groups, so a group member can share your child’s words with it. Treat anything said in a group as potentially shared.

Is AI built into apps safer than a standalone chatbot?

Not necessarily. Several embedded assistants received Common Sense’s worst rating, including Google’s AI Mode, Meta AI and Grok. Embedded AI is also harder to see, because kids don’t think of it as a separate tool.

Will laws force apps to turn off AI for kids?

Possibly in Europe. A European Commission proposal reported in September 2026 would require embedded chatbots to be off by default for minors. It is not final. No US federal rule does this; state laws focus on companion features instead.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

---

Sources

1. Youth AI Safety Institute. (2026, July 14). “Google Search: AI Overview & AI Mode Risk Assessment.” Common Sense Media. https://institute.commonsensemedia.org/risk-assessments/google-search
2. Youth AI Safety Institute. (2026). “Risk Assessments.” Common Sense Media. https://institute.commonsensemedia.org/risk-assessments
3. Mann, S. (2026, October 27). “Almost half of tweens have used an AI chatbot or companion, new research finds.” Common Sense Media Youth AI Safety Institute. https://institute.commonsensemedia.org/tween-teen-census-2026
4. Snapchat Support. (n.d.). “How do I limit my teen’s access to My AI?” Snap Inc. https://help.snapchat.com/hc/en-us/articles/22628526282772-How-do-I-limit-my-teen-s-access-to-My-AI
5. WhatsApp. (2026). “Privacy questions.” WhatsApp. https://www.whatsapp.com/privacyquestions
6. SBT News. (2026, August 28). “ANPD afirma que Meta discutirá proteção infantil no Brasil.” SBT News. https://sbtnews.sbt.com.br/noticia/mundo/anpd-afirma-que-meta-discutira-protecao-infantil-no-brasil
7. Office of the Governor of New York. (2026, October 28). “Governor Hochul Signs Nation-Leading Legislation to Protect Kids From AI Chatbots and to Ensure Baby Monitor Safety.” NY.gov. https://www.governor.ny.gov/news/governor-hochul-signs-nation-leading-legislation-protect-kids-ai-chatbots-and-ensure-baby
8. Raya, A. (2026, September 17). “La UE marca las nuevas normas: los chatbots de IA y los videojuegos online, también limitados para adolescentes.” El Español. https://www.elespanol.com/omicrono/20260917/ue-marca-nuevas-normas-chatbots-ia-videojuegos-online-limitados-adolescentes/1003744387087_0.html