🤖 Meta Halts Teen Access to AI Characters

Meta is hitting pause on teen access to its AI characters. After months of backlash over risky and inappropriate interactions, the company says it will temporarily block teenagers worldwide from chatting with its AI bots, a move that signals growing pressure on tech giants to put safety first as AI races ahead. Let’s dive in!

In today’s insights:

LATEST DEVELOPMENTS

Evolving AI: Meta is pausing teens’ access to AI characters after safety concerns and policy scrutiny.

Meta announced it will temporarily stop teens worldwide from chatting with its AI character bots after months of criticism over unsafe interactions. Internal policy errors had allowed “sensual” conversations with minors, triggering media reports and legal attention. The company says updated controls will block topics like self-harm and eating disorders and add parental oversight. The ban will cover teen accounts and users flagged as underage by Meta’s age prediction systems, though access to the standard Meta AI bot will continue.

Meta hitting pause on teen access to its AI characters is a signal that “companion” chatbots are sliding into the same safety lane as social apps, with real guardrails and real oversight. Meta says the block is rolling out in the coming weeks and stays in place until a new teen version ships with parental controls and tighter content limits, including a PG-13 style standard. Regulators are circling too, with the FTC’s 2025 inquiry into AI companion bots pushing companies to show how they test risks for kids. For families, it’s a quick reduction in awkward or harmful chats; for AI product teams, it’s a reminder that age checks, logging, and parent-visible settings are becoming table stakes before features can scale.

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide

Evolving AI: CEOs warn that fast AI adoption may replace roles faster than new work appears

Business leaders at Davos voiced concern that AI could reshape employment sooner than many expect. Dan Schulman said systems are close to moving from support to replacement, with general intelligence forecast within a few years and robots soon after. Data from Cognizant shows rising exposure across office and manual roles. Firms debate new org shapes, reskilling, and public policy as hiring patterns shift and graduate roles thin.

The talk in Davos is lining up with a wider pattern: AI is moving from “nice helper” to “staffing lever,” and the squeeze shows up first in entry level office work where routine analysis, drafting, and checking can be done fast and cheap. Cognizant’s latest study of 18,000 tasks flags exposure scores rising about 9% a year, so roles that felt safe a year ago look shakier now. Add in Dario Amodei’s warning that unemployment could hit the 10% to 20% range if companies automate faster than they retrain, and you get why graduate hiring and junior career paths feel tense. For workers, the near term win is learning the “AI supervisor” parts of the job: write better briefs, spot bad outputs, own the final call, and document impact so your value is visible. For leaders, the practical play is redesigning workflows and training plans now, before org charts quietly hollow out.

We tipped this documentary before, and now it’s exploded past 300 million views. The Thinking Game takes you inside DeepMind’s journey from beating humans at Go to solving protein folding and chasing AGI, with rare behind-the-scenes access to how modern AI breakthroughs actually happen. If you haven’t watched it yet, this is one of those “how did I miss this?” moments.

Evolving AI: Sparkli uses generative AI to turn children’s questions into daily interactive learning expeditions.

Sparkli is an AI-powered learning app for children aged 5–12 created by three former Google employees. It delivers daily “expeditions” that combine narrated stories, images, audio, quizzes, and games based on preset themes or a child’s own questions. Tested in more than 20 schools and reaching over 100,000 students through partners, the app supports teachers with progress tracking and homework tools. Backed by $5 million in pre-seed funding, Sparkli plans a wider release for families by mid-2026.

When gen AI moves from chat boxes to full “expeditions” with audio, images, quizzes, and teacher dashboards, it starts fitting into how school actually works: a quick hook at the start of class, a guided recap for homework, and progress tracking without extra spreadsheets. Sparkli is riding that shift as it pilots in schools and lines up a broader rollout for families. The timing is tight too, since policymakers and child-safety groups are pushing harder on kids’ AI use, privacy, and age-appropriate design in Europe. For teachers and parents, the practical question becomes simple: can an app like this give kids a safer, structured way to explore curiosity than open-ended web search and random video feeds?

Turn spoken explanations into accurate, formatted financial copy for reports and investor comms. Wispr Flow saves editing time and keeps messaging consistent. Try Wispr Flow for finance.

Try Wispr Flow

QUICK HITS

📷 Google Photos rolling out AI-powered ‘Me Meme’ generator

📉 Intel CEO ‘disappointed’ as 13% share drop indicates AI data centre struggle

🫢 Is China quietly winning the AI race?

🤖 GitHub Releases Copilot-SDK to Embed Its Agentic Runtime in Any App

💊 How AI and robots help ease pressure on pharmacies

🏆 ‘The Silicon Gaze’: ChatGPT rankings skew toward rich Western nations, research shows

 *partner link