If you’re thinking about how to apply AI to your business, especially using ChatGPT — which still leads the pack when it comes to practical LLMs for organizations — they just rolled out some serious upgrades inside their “Projects” feature. Here’s a quick breakdown (and why it matters for us operators): 1️⃣ Deep Research Mode (Now inside Projects) You can now run what OpenAI calls “Deep Research” directly inside your projects. This mode allows ChatGPT to pull from your project’s internal context (chats, files, instructions) and supplement it with fresh, live data from the web. Think of it like AI-powered research reports — much deeper than a casual chat. Before this, you could only run Deep Research outside of your projects — now it fully leverages your internal data too. 2️⃣ Voice Mode Now Supports Projects ChatGPT’s Voice Mode (which frankly blows away every other AI assistant in terms of real-time conversation) is now fully compatible with Projects. You can literally walk, drive, or work hands-free while having an intelligent conversation about your files, chats, and data inside a project. Imagine reviewing your quarterly reports while on your morning walk, or brainstorming new product features while commuting — all conversationally. 3️⃣ Project-Level Memory Across Chats Previously, memory in ChatGPT was limited. Now, memory works across all chats within a project. If you mention something in one chat, that knowledge can carry over and be referenced automatically in other chats inside the same project. (Note: Only available for Plus/Pro users) 4️⃣ Mobile Uploads + Model Switching You can now upload files and switch models directly from the ChatGPT mobile app — a big unlock for anyone who prefers mobile-first workflows. These features were previously desktop-only. 5️⃣ Seamless Project Creation from Existing Chats If you’re mid-conversation and decide “this should be a project” — you can now retroactively convert any chat into a project or attach it to an existing one. No need to start over.