Subvolts take OpenAI tunes GPT-5.1 for personality, balancing intelligence with emotional response. Dec 2, 2025 - Developers
- Learn about techniques for controlling AI model outputs and user interactions with models like ChatGPT.
- Business
- Understand how AI "personality" and responsiveness are developed, impacting user experience and AI adoption.
- Everyday users
- Discover how AI models are being tuned to be more helpful, emotionally aware, and flexible in their responses.
This episode features OpenAI researchers discussing concrete details about GPT-5.1's development and behavior tuning.
Shaping Model Behavior in GPT-5.1 - Episode 11
Audio streams from OpenAI. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
In The OpenAI Podcast, researcher Christina Kim and product manager Laurentia Romaniuk discuss shaping GPT-5.1's behavior. They explore how OpenAI balances AI's "IQ" and "EQ," aiming for models that offer both intelligence and emotional responsiveness. The conversation covers user feedback, model steerability, and the blend of art and science in AI development.
This episode explains how AI like ChatGPT gets its "personality." Researchers discuss making the AI smarter and more emotionally aware, giving users more control over how it acts. It's about blending logic and feeling to make AI more helpful.
This episode offers insight into how OpenAI engineers AI "personality" and user-facing flexibility. It's for anyone interested in the practical application of AI research, particularly how models are tuned beyond raw capability. The discussion on steerability and user feedback connects directly to ongoing efforts across frontier labs to make AI more controllable and aligned with human preferences.
- HostAndrew Mayne
- GuestChristina KimResearcher, OpenAI
- GuestLaurentia RomaniukProduct Manager, OpenAI
- Episode Title: Shaping Model Behavior in GPT-5.1 - Episode 11
- Show: The OpenAI Podcast
- Length: 29 minutes
- Guests: Christina Kim, Laurentia Romaniuk
- Topics Covered: GPT-5.1 goals, model switcher, user feedback, emotional intelligence, model personality, steerability, bias, uncertainty, memory in ChatGPT
What does it mean for an AI model to have "personality"? Researcher Christina Kim and product manager Laurentia Romaniuk talk about how OpenAI set out to build a model that delivers on both IQ and EQ, while giving people more flexibility in how ChatGPT responds. They break down what goes into model behavior and why it's an important, but still imperfect blend of art and science.
Chapters:
- 00:00:43 — GPT-5.1 goals and the shift to reasoning models
- 00:02:18 — Differences between GPT-5 and GPT-5.1
- 00:04:55 — Unpacking the model switcher
- 00:07:24 — Understanding user feedback
- 00:08:27 — Measuring progress on emotional intelligence
- 00:10:02 — What is model personality?
- 00:14:25 — Model steerability, bias, and uncertainty
- 00:21:59 — Advantages of memory in ChatGPT
- 00:25:27 — Looking ahead and advice for getting the most out of models
Hosted on Acast. See acast.com/privacy for more information.
- IQ and EQ
- Christina Kim and Laurentia Romaniuk
- Intelligence Quotient (IQ) and Emotional Quotient (EQ) refer to a model's ability to reason and understand/express emotions.
- Model steerability
- Christina Kim and Laurentia Romaniuk
- The ability to guide or direct an AI model's behavior and outputs in a desired way.
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
29 minInside image generation’s Renaissance moment - Episode 19
WithAdele Li, Kenji Hata
2 h 01US vs. China: Why Trust Will Win the AI Race | GPT-5.2 & Anthropic IPO w/ Emad Mostaque, Salim Ismail, Dave Blundin & Alexander Wissner-Gross | EP #214
WithEmad Mostaque, Salim Ismail, Dave Blundin, Alexander Wissner-Gross
1 h 04
30 minScience, Innovation, and Economic Growth: OpenAI’s Ronnie Chatterji
WithRonnie Chatterji
Questions people ask
- What are the main goals for GPT-5.1?
- The episode discusses a shift towards reasoning models and aims to build a model that delivers on both IQ and EQ, offering users more flexibility in ChatGPT's responses.
- How does OpenAI approach "model personality"?
- It involves an imperfect blend of art and science, focusing on steerability, bias, and uncertainty to shape how the model behaves and responds.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to OpenAI.