Google’s Gemini Fails Key Child Safety Tests, Says Watchdog

✨TodayOnAI’s Daily Drop

• Google’s Gemini Fails Key Child Safety Tests, Says Watchdog

• Snapchat Launches AI-Powered Lens for Custom Image Generation

• Koah Raises $5M to Bring Ads to the AI Chatbot Ecosystem

• 💬 Let’s Fix This Prompt

• 🧰 Today’s AI Toolbox Pick

Google

Google’s Gemini Fails Key Child Safety Tests, Says Watchdog

🚀 TodayOnAI Insight: Common Sense Media has rated Google’s Gemini AI as a high risk for kids and teens, warning that despite some safeguards, the platform still exposes young users to harmful content and lacks age-appropriate design. With Apple reportedly eyeing Gemini for Siri integration, the stakes for AI child safety are rising fast.

🔍 Key Takeaways:

• Adult AI under the hood: Gemini’s “Under 13” and “Teen Experience” tiers use largely the same models as the adult version, with only surface-level safety tweaks.

• Risky content exposure: Tests showed Gemini still delivers unsafe advice and sensitive information on sex, drugs, and mental health, even with filters enabled.

• Lack of developmental nuance: The platform fails to tailor interactions to users' age and maturity, ignoring critical differences in guidance needs.

• Safety concerns amid real-world harm: The report follows recent lawsuits linking AI use to teen suicides, including a wrongful death case involving ChatGPT.

• Google’s response: The company acknowledged issues in Gemini’s responses, claimed improvements are ongoing, and emphasized red-teaming and expert reviews.

💡 Why This Stands Out: This assessment signals a deeper issue in how major AI models are retrofitted for youth rather than built for them. With tech giants like Apple potentially embedding LLMs like Gemini into core products, the absence of child-first design poses real risk. If AI is to serve younger users safely, companies will need to move beyond patches and filters—and start from principles of age-appropriate, developmental design.

Snapchat

Snapchat Launches AI-Powered Lens for Custom Image Generation

🚀 TodayOnAI Insight: Snapchat is rolling out its first open-prompt generative AI Lens, Imagine Lens, enabling subscribers to create and edit images using text-to-image prompts. The move deepens Snap’s integration of generative AI within its AR ecosystem.

🔍 Key Takeaways:

• First-of-its-kind AI Lens: Imagine Lens lets users generate images from custom text prompts, like “turn me into an alien” or “create a grumpy cat.”

• Flexible prompt editing: Users can tap the caption bar to modify prompts, or choose from pre-loaded creative scenarios.

• Exclusive access: Available only to Snapchat+ Platinum ($15.99/month) and Lens+ ($8.99/month) subscribers.

• AI model stack: Snap says the feature uses both in-house and third-party models but hasn’t confirmed if its new mobile image model powers the Lens.

• AI strategy expands: Follows recent releases like AI video Lenses and the standalone Lens Studio tool for AR creation.

💡 Why This Stands Out: Snap is leaning further into generative AI to enhance creativity and drive premium subscriptions. While rivals focus on chatbots and productivity tools, Snap is staking out the intersection of AI, mobile creativity, and AR. Will its visual-first approach to AI carve out a distinct space in a crowded landscape?

Koah

Koah Raises $5M to Bring Ads to the AI Chatbot Ecosystem

🚀 TodayOnAI Insight: AI monetization is still murky for most developers, but Koah is betting big on ads. The startup just raised $5 million to build a native ad platform tailored to consumer AI apps—especially those reaching global users who won't pay subscription fees.

🔍 Key Takeaways:

• Koah raised a $5M seed round led by Forerunner, with backing from South Park Commons and AppLovin’s co-founder.

• Its ad tech targets the “long tail” of AI apps—not ChatGPT itself—serving users in cost-sensitive regions like Latin America.

• Early results show Koah delivering 7.5% clickthrough rates and $10K+ monthly earnings for app partners like Luzia and Liner.

• Ads appear contextually during chats (e.g., UpWork ad during startup advice) and are labeled as sponsored content.

• Koah claims to outperform legacy ad networks like Admob and AppLovin in both effectiveness and user experience.

💡 Why This Stands Out: As subscription fatigue sets in, especially outside wealthier markets, Koah’s model signals a shift toward advertising as a default monetization layer for consumer AI. Its traction suggests that smart, contextual ads can work in chat interfaces—without degrading user experience. The real test: Can ads in AI finally become as native and profitable as they are on the rest of the internet?

✨ See how a simple prompt upgrade can unlock better AI output.

🔹 The Original Prompt

At first glance, this prompt might seem okay. But it's too broad — and that limits the quality of AI-generated results. Let’s improve it using prompt engineering best practices.

✅ The Improved Prompt

💡 Why It's Better

• Specific audience: Targets decision-makers in mid-sized companies.

• Contextual focus: Emphasizes emerging tech and practical solutions.

• Actionable output: Requests summaries and titles to spark execution.

• Tone and style: Guides the type of content (insightful, engaging, relevant).

🛠️ Learn how to adapt this prompt for SaaS, AI tools, dev teams & more →
Read the full PromptPilot breakdown

💡 Bonus Tool: Want to generate and master prompts instantly?
👉 Try PromptPilot by TodayOnAI (Free to use)

📰 More from the AI World

• What is Mistral AI? Everything to know about the OpenAI competitor

• OpenAI hires the team behind Xcode coding assistant Alex

• Personalized AI companion app Dot is shutting down Sarah Perez M

• Attorneys general warn OpenAI ‘harm to children will not be tolerated’

🧰 Today’s AI Toolbox Pick

• ⚙️AlloBOT (Customer Support Tool): An AI-powered phone service that enhances customer support.

• 🍉Watermelon (Chatbot Tool): Builds a personal GPT-4 chatbot with your data.

• 🧨Moodboard Creator (Content Tool): Helps designers spark their creativity.