Hello, this is Ryuta Hamamoto from TIMEWELL.
ChatGPT has gained "eyes." In December 2024, OpenAI added video calling and screen sharing to Advanced Voice Mode, introducing the dimension of "vision" to AI communication that had previously been limited to text and voice. This article provides a comprehensive overview of how these features work, how to use them in business, and the evolution of multimodal AI through 2026.
Video Calling: AI That Understands Video in Real Time
The video calling feature announced by OpenAI on December 12, 2024 is a groundbreaking update that enables camera video input in ChatGPT's Advanced Voice Mode. When a user points their smartphone camera at something, ChatGPT analyzes the video feed in real time and responds by voice.
Key Features
| Feature | Details |
|---|---|
| Real-time video recognition | Instantly identifies objects, people, and text captured by the camera |
| Voice integration | Continues natural voice conversation while viewing video |
| Memory capability | Remembers names of people introduced during conversation |
| Multilingual support | Responds in 50+ languages, including emotional tone |
| Response speed | ~232 milliseconds (comparable to human conversation speed) |
In a demonstration, ChatGPT recognized a coffee maker through the camera and provided step-by-step guidance on how to make hand-drip coffee — from preparing the filter to the bloom and pouring technique. The system also demonstrated integrated vision and memory, instantly answering questions like "What was the name of the colleague wearing the Santa hat?"
How to Use It
- Install the latest version of the ChatGPT mobile app
- Tap the "Advanced Voice Mode" button in the lower right of the chat screen
- Tap "Video" to start a video call
- Point the camera at your subject and talk with ChatGPT
Screen Sharing Expands AI Use Cases
Added alongside video calling, screen sharing lets users share their smartphone screen with ChatGPT. Rather than just pointing a camera around, ChatGPT directly analyzes the apps and content on your screen and provides real-time advice.
Screen Sharing Use Cases
- Message reply assistance: Share your messaging app screen to get contextually appropriate reply suggestions
- Code review: Share a programming screen to identify errors and get code improvement advice
- Document review: Display presentation slides or spreadsheets to receive improvement feedback
- Operation guidance: Show your screen when you're unsure how to use an app to get step-by-step instructions
Usage is the same as video calling — simply select "Share Screen" from Advanced Voice Mode.
Looking for AI training and consulting?
Learn about WARP training programs and consulting services in our materials.
GPT-4o: The Technical Foundation Behind Multimodal Capabilities
These features are powered by OpenAI's GPT-4o model. GPT-4o uses a "natively multimodal" design that processes text, voice, images, and video in an integrated manner.
Comparison With Previous Models
| Aspect | Previous Models | GPT-4o |
|---|---|---|
| Voice processing | Speech → text conversion → processing → text → speech synthesis | Directly processes speech and outputs speech directly |
| Video understanding | Static images only | Real-time video stream support |
| Response speed | Several seconds of lag | ~232 milliseconds (near real-time) |
| Emotion recognition | Inference from text content | Directly understands voice tone, speed, and emotion |
By eliminating the intermediate text conversion step, latency was significantly reduced, achieving response speeds close to natural human conversation.
Practical Business Applications
Video calling and screen sharing can be applied across a wide range of business scenarios.
1. Remote Support for On-Site Work
In manufacturing and construction environments, workers can point a camera at equipment to troubleshoot. AI analyzes the situation from the video and provides step-by-step voice guidance through the repair process.
2. Real-Time Review of Sales Materials
By sharing proposals or presentation slides via screen share, you can instantly receive feedback on structural improvements and how to present data.
3. Multilingual Communication
With support for 50+ languages, the system can serve as a real-time interpreter when communicating with overseas partners.
4. Education and Training
For new employee training, you can build a system where AI provides individualized instruction while viewing the actual work screen or work environment.
2026 Update: The Evolution of Multimodal AI
From 2025 through 2026, ChatGPT's multimodal capabilities have continued to evolve significantly.
Key Updates
- Seamless voice conversation integration (November 2025): Voice conversations can now be conducted directly within existing chat threads, with text, images, maps, and voice displayed in parallel in real time
- Opened to free users (February 2025): Advanced Voice Mode preview became available to free users
- GPT-5.2 release (December 2025): Became the default model across all plans, with further improvements in multimodal processing
- GPT-5.3-Codex (February 2026): Evolved beyond code generation into a general-purpose work agent
As of 2026, ChatGPT holds a 64.5% market share (down from 86.7% in early 2025), with Google's Gemini growing to 21.5%, reflecting the rapid expansion of the multimodal AI market as a whole.
How TIMEWELL WARP Supports Enterprise AI Adoption
The evolution of multimodal AI has greatly expanded the possibilities for enterprise AI use. At the same time, many companies report concerns such as "we're not sure how to implement this in our organization" or "we're worried about security."
TIMEWELL's AI consulting service "WARP" addresses these challenges.
- WARP: Expert-guided support from AI strategy formulation through implementation
- WARP NEXT: Develop and implement an AI adoption plan for existing business operations step by step
- WARP BASIC: A training program to learn the fundamentals of AI utilization
Former DX and data strategy specialists from major enterprises propose AI adoption strategies tailored to your company's situation — including the latest multimodal AI technology.
Summary
- ChatGPT's Advanced Voice Mode now features video calling and screen sharing, enabling "visual dialogue" with AI
- GPT-4o's natively multimodal design enables near real-time responses of approximately 232 milliseconds
- Supports 50+ languages with natural conversation that captures emotion and tone
- Applicable across a wide range of business scenarios including on-site support, sales, and training
- In November 2025, voice conversations were seamlessly integrated into chat threads
- With GPT-5.2/5.3 emerging in 2026, the multimodal AI market is transitioning from "experimental phase" to "full adoption"
- When unsure about AI adoption, consulting with TIMEWELL WARP specialists is the fastest path forward
References
- OpenAI Official - Day 6: Advanced voice with video & Santa mode
- ChatGPT Release Notes - OpenAI Help Center
- Advanced Voice Mode FAQ - OpenAI Help Center
- ChatGPT gets screensharing and real-time video analysis - VentureBeat
- Introducing GPT-5.2 - OpenAI
Related Articles
- The Reality of a Part-Time Employee Who Worked Full-Time, Took Two Maternity Leaves, and Changed Her View of Work | TIMEWELL
- Before Paternity Leave — What You Absolutely Must Do to Take Leave Even During a Busy Period
- Pursuing a Hands-On Architecture Firm: Finding My Own Way as the 5th Generation of a Construction Company | Fujita Construction
