Subvolts take See how AI image generation is evolving from DALL-E to creative assistants. May 14, 2026 - Developers
- Learn about the latest image generation features and potential integrations with tools like Codex.
- Business
- Discover new use cases and trends in AI image generation for marketing and creative content.
- Everyday users
- Understand how AI image tools are improving and becoming more capable creative assistants.
This interview provides a detailed look at recent advancements in AI image generation from a leading lab.
Inside image generation’s Renaissance moment - Episode 19
Audio streams from OpenAI. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
This episode of The OpenAI Podcast features Product lead Adele Li and researcher Kenji Hata discussing the evolution and impact of image generation since the launch of Images 2.0. They explore new use cases, trends, and advancements like improved text rendering, photorealism, and character consistency. The conversation also touches on the future of image generation models as creative assistants, including their integration with coding tools.
AI can now create over 1.5 billion images weekly in ChatGPT. This episode explains how these tools got better, from making realistic pictures to understanding text and keeping characters consistent. It's like having a digital artist that's constantly improving.
This episode offers insight into the rapid progress of AI image generation, a key area for all frontier labs. Listeners will learn about the practical applications and creative potential unlocked by recent model improvements. It's for anyone tracking AI's creative capabilities, from developers building on these tools to businesses exploring new marketing avenues. The discussion on creative agents and future directions directly relates to ongoing work at Google Gemini, Anthropic, and Meta AI, highlighting the competitive landscape.
- HostAndrew Mayne
- GuestAdele LiProduct lead, OpenAI
- GuestKenji HataResearcher, OpenAI
- Episode Title: Inside image generation’s Renaissance moment - Episode 19
- Show: The OpenAI Podcast
- Length: 29 minutes
- Weekly Image Generation: Over 1.5 billion images a week in ChatGPT
- New Capabilities: Better text rendering, photorealism, multilingual support, world knowledge, aspect ratios, character consistency
People are generating over 1.5 billion images a week in ChatGPT. In this episode, Product lead Adele Li and researcher Kenji Hata share some of the new use cases and trends since the launch of Images 2.0. Together with host Andrew Mayne, they trace the progress from the early DALL-E days and dive into the latest capabilities, including better text rendering, photorealism, multilingual support, world knowledge, aspect ratios, and character consistency. They also explore what comes next as image generation models evolve into more capable creative assistants.
Chapters
00:36 How Adele and Kenji came to work on Images
02:27 Images 2.0 launch reception
05:25 Productivity use cases and and 360 images
09:34: Viral trends, authenticity, and imperfection
10:51 Training breakthroughs and photorealism
14:06 Evals, prompting, and creative control
22:16 Creative agents and what comes next
22:27 Images + Codex
28:08 Prompt tips
Hosted on Acast. See acast.com/privacy for more information.
- Images 2.0
- Kenji Hata
- OpenAI's latest version of its AI image generation technology.
- photorealism
- Kenji Hata
- The ability of AI to create images that look like real photographs.
- character consistency
- Kenji Hata
- AI's ability to keep a character looking the same across multiple generated images.
- Codex
- Kenji Hata
- An AI model that translates natural language into code.
41 minWhat racing reveals about working with AI - Episode 22
WithJoyce Ruffell, Chase Holden
44 minWhy Tejal Patwardhan stopped underestimating the models - Episode 21
WithTejal Patwardhan
41 minHow a reasoning model cracked an 80-year-old math problem - Episode 20
WithAlexander Wei, Hongxun Wu, Lijie Chen
38 minWhy AI needs a new kind of supercomputer network - Episode 18
WithMark Handley, Greg Steinbrecher
1 h 17
1 h 14OpenAI's Yann Dubois: Why AI Progress Suddenly Feels Real
WithYann Dubois
34 minMusk v Altman: Much ado about nothing
WithLiz Lopatto
52 minHermes Agent: Agents that grow with you
WithJeffrey Quesnelle
Questions people ask
- How many images are people generating weekly with ChatGPT?
- People are generating over 1.5 billion images a week in ChatGPT.
- What new capabilities does Images 2.0 offer?
- Images 2.0 offers better text rendering, photorealism, multilingual support, world knowledge, aspect ratios, and character consistency.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to OpenAI.