The Big Story

ChatGPT is getting closer to letting you make and send custom stickers directly inside WhatsApp. The development comes from reports that the chatbot could soon generate and deliver AI stickers straight into WhatsApp conversations. Instead of searching for existing stickers or drawing your own, you could describe what you want—like a cartoon version of your friend making a silly face—and the AI would generate it on the spot. Think of it like having a tiny artist living inside your messaging app who instantly turns your words into shareable images. This builds on how current AI image tools already create pictures from text prompts, but the new step is seamless delivery straight into chats without extra apps or downloads. The feature would let users stay inside one conversation flow rather than switching between tools. For teens and students, this could make group conversations more fun and personal, whether you're reacting to memes, planning events, or just goofing around with friends. It also opens creative doors for school projects or online communities where quick visuals help ideas land better. The change matters because it removes friction between thinking of something funny and actually sharing it. You might soon try this by opening ChatGPT on your phone, describing a sticker idea, and seeing if the option to send it to WhatsApp appears in the menu. Source: Google News

Explain Like I'm 14

You know how when you text a friend, your phone sometimes finishes your sentence based on what you've said before? Now picture that same idea, but the AI has read millions of stories, jokes, and conversations from books and the internet. It breaks every piece of writing into small chunks called tokens, then predicts which token should come next to keep the story flowing naturally. Step by step, it builds sentences the same way you might keep adding details to a story you're telling out loud. In one study, more than 2,500 participants read short stories without knowing the author. They rated the AI-generated versions higher for creativity and flow than the human-written ones. The moment they learned a machine created the text, those scores dropped because expectations changed. So the technology isn't copying one specific story—it's using patterns from everything it's seen to invent something new that still feels human. Source: the-decoder.com

Cool Stuff & Try This

ChatGPT now supports bold and italic text in your prompts You can now type italic or bold directly when asking ChatGPT questions, just like on Reddit or in notes apps. This helps you emphasize parts of your request so the AI focuses on exactly what matters. It's useful for school assignments or creative writing where you want clear structure. The change appeared in the ChatGPT interface after some users upgraded plans. Open the ChatGPT app or website, start a new chat, and try typing a prompt with bold around the most important instruction to see if the response improves. Source: reddit.com

Turn off Gemini suggestions in Gmail and Google Docs Google's AI helper sometimes pops up with writing ideas while you're emailing or working on documents. If you prefer working without those prompts, you can switch them off in the settings. This keeps your workspace simpler and gives you full control. The feature appears as toolbars and prompts inside the apps. Go into Gmail or Google Docs on your computer or phone, open the settings menu, and look for the Gemini or AI features toggle to turn suggestions off. Source: wired.com

Four easy ways to use AI while using less energy Running AI tools uses electricity, and small choices add up over time. Experts suggest using shorter prompts, picking text-only answers when you don't need images, and closing chats when you're done. You can also try asking the AI for quick answers instead of long back-and-forth conversations. These steps reduce the power drawn by data centers. Next time you open ChatGPT or Gemini, test one of these by keeping your request short and seeing the result. Source: livescience.com

Quick Bits

Samsung support chat accidentally pasted its own AI prompt A customer service worker copied the hidden instructions meant for the AI right into the chat with a user. It showed exactly how the system was told to respond. Source: reddit.com

Claude Code sessions can now talk to each other across terminals On macOS and Linux, separate Claude Code sessions running at the same time can send messages and share status updates. This lets multiple AI coding helpers coordinate without the user copying information between windows. Source: the-decoder.com

💬 Reply to this email — Patrick reads every one.

Share: X · LinkedIn · WhatsApp

Forwarded this email? Subscribe here — it's free.