AI Tools Of The Week: This week’s AI tools go beyond answering questions; they do the work for you. From building interactive tools directly inside a chat, to preparing meeting briefs by pulling data across apps, to converting dense research into video summaries, these features reflect a shift toward AI as an active collaborator rather than a passive assistant.
Google Gemini Mini-Apps (Powered by Opal)
What problem does Mini-Apps in Gemini solve?
Most AI chatbots stop at giving advice or explanations. When you ask for a budget calculator, project tracker, or comparison table, you still have to copy responses into spreadsheets or recreate them in other tools. The AI understands the task, but the execution remains manual.
This gap between insight and action creates friction and slows productivity.
Gemini’s new mini-apps, built using Google’s experimental Opal framework, close this gap by generating fully interactive tools directly inside the chat, ready to use, edit, and share instantly.
How to access
Go to gemini.google.com → Click the beaker icon (top right) → Enable Gems with mini apps
What Mini-Apps in Gemini Help You Do
- Create interactive tools instantly: Build calculators, planners, trackers, workflows, or games without leaving Gemini.
- Prototype ideas quickly: Test functional app concepts before committing to full-scale development.
- Collaborate more effectively: Share tools that teammates can interact with, not just static text outputs.
Example
You’re launching a new product and want a structured workflow for your marketing team.
- Step 1: Click New Gem to start creating a mini-app
- Step 2: Describe your need: “Create an app that evaluates product features, generates marketing content ideas, builds vision boards, and outlines next steps.”
- Step 3: Gemini builds the app automatically, including sections for product details, content generation, visual concepts, and outputs
- Step 4: Your team uses the app immediately to input data and generate results
- Step 5: Click Remix to refine steps or customise the workflow further
What makes Gemini mini-apps stand out?
- No technical expertise required: Apps are built through conversation
- Everything in one place: No exporting or switching platforms
- Instant iteration: Modify functionality by simply asking
ChatGPT Connectors
What problem does ChatGPT Connectors solve?
Preparing for meetings often involves frantic app switching, checking email threads for context, scanning calendars for attendees, searching Drive for files, and mentally stitching everything together. This routine consumes time and mental energy before the meeting even begins.
ChatGPT Connectors eliminate this friction by allowing the model to directly access Gmail, Google Calendar, Google Drive, and other supported apps, bringing all relevant information into a single, structured response.
How to access
Click your ChatGPT profile → Settings → Connectors → Enable Gmail, Google Calendar, and Google Drive
What ChatGPT Connectors help you do
- Prepare smarter: Automatically pull attendee details, email history, and related documents.
- Simplify collaboration: Create a single source of truth when working with shared calendars and drives.
- Extend beyond meetings: Use connectors for research, writing, planning, and brainstorming.
Example
You have a strategy meeting scheduled tomorrow.
- Step 1: Enable connectors via the prompt box (‘+’ → Connector apps)
- Step 2: Ask ChatGPT: “Review my next meeting and prepare a brief with attendees, email context, related Drive files, and suggested discussion points.”
- Step 3: ChatGPT generates a structured briefing, attendee profiles, key email summaries, file links, and a draft agenda. You can add: “Highlight potential risks based on past discussions,” and it will flag concerns.
What makes Connectors special?
- Unified view: Gmail, Calendar, and Drive in one interface
- Reduced cognitive load: No repetitive searching
- Proactive insights: Anticipates issues, not just information
Google NotebookLM's Video Overview
What problem does Video Overview solve?
Learning from dense research materials is challenging, not just because of volume, but because people learn differently. Text-heavy documents overwhelm visual learners, while others struggle without a clear structure or narration.
Traditional study methods force everyone into the same reading-heavy workflow, often leading to poor retention and information overload.
NotebookLM’s Video Overview feature addresses this by transforming large sets of documents into multimodal video presentations that combine visuals and narration.
How to access
Available at notebooklm.google.com
What Video Overview helps you do
- Create presentation-ready content: Generate professional videos from raw research materials.
- Synthesise multiple sources: Combine papers, articles, and notes into one coherent narrative.
- Support different learning styles: Engage visual and auditory learners simultaneously.
Example
You’re researching AI ethics and have gathered multiple papers and articles.
- Step 1: Upload all sources to a NotebookLM notebook
- Step 2: Select Video Overview
- Step 3: NotebookLM produces a structured video with slides, charts, and narration
- Step 4: You absorb key ideas and connections far faster than reading each source individually
What makes Video Overview stand out?
- Multimodal learning: Visuals plus narration improve retention
- Automatic synthesis: AI identifies themes and relationships
- Professional output: Presentation-ready videos without design skills
Known limitations: Processing takes longer than audio summaries, visual customisation is limited, and results are best with text-heavy inputs.
(Bindra & Magazine are the co-founders of AI&Beyond)
Disclaimer: The tools and analysis featured in this section demonstrated clear value based on our internal testing. Our recommendations are entirely independent and not influenced by the tool creators.