Sora 2 is Here

In partnership with

Welcome to VP Land! Sora 2 is here, along with a new social app from OpenAI for AI generated videos. We've got the details of what's new in the model. It feels like OpenAI has entered the AI video model chat again - but personally, I'm looking for new ways of creating video that aren't just text or image inputs.

In the last poll, most of you agreed that the biggest hurdle for high-quality studio AI film generation is untangling legal rights for existing actor likenesses/voices. Check out today's poll below.

In today's edition:

Sora 2
An AI 'actress' might get talent representation
Wan 2.5 Preview Introduces Audio-Driven Video Generation
Tencent Debuts HunyuanImage 3.0

Sora 2 is Here. Plus a Social App

OpenAI released Sora 2, a major upgrade to its video generation model that the company describes as jumping "straight to what we think may be the GPT-3.5 moment for video." The company also just launched a new social iOS app called Sora, powered by the Sora 2 model.

The most significant improvements focus on physics accuracy and controllability:

Better object permanence and physics modeling - The original Sora would "morph objects and deform reality" to execute prompts (like making a missed basketball shot teleport to the hoop), while Sora 2 actually models realistic outcomes like rebounds off backboards
Cameos feature - Users can upload a short video and audio recording to inject themselves into any generated scene with "remarkable fidelity," maintaining accurate appearance and voice
Multi-shot consistency - The model can follow complex instructions across multiple shots while maintaining world state, handling realistic, cinematic, and anime styles
Synchronized dialogue and sound effects - As a versatile video-audio generation system, it can produce realistic background soundscapes, speech, and sound effects with impressive detail

Sora 2 launches through the new social iOS app Sora, with social features built around the cameos capability. The service starts free with "generous limits" in the US and Canada, with ChatGPT Pro users getting access to a higher-quality Sora 2 Pro model. An API release is also planned.

In a separate blog post, CEO Sam Altman called this the "ChatGPT for creativity" moment, noting the team's focus on avoiding addictive social media patterns through features like natural language feed control and periodic wellbeing checks.

What matters: The physics improvements address real limitations that made previous AI video feel obviously artificial. The cameos feature could be genuinely useful for quick creative projects, though the social app approach feels like uncharted territory for a video generation tool.

SPONSOR MESSAGE

You Don’t Need to Be Technical. Just Informed

AI isn’t optional anymore—but coding isn’t required.

The AI Report gives business leaders the edge with daily insights, use cases, and implementation guides across ops, sales, and strategy.

Trusted by professionals at Google, OpenAI, and Microsoft.

👉 Get the newsletter and make smarter AI decisions.

Lead with AI—no coding needed

Is Tilly Norwood an 'AI Actress' or Just an Influencer?

Talent agencies are reportedly in talks to represent Tilly Norwood, an AI-generated actress created by Eline Van der Velden's studio. According to Deadline, this would mark one of the first times a talent agency signs an AI-generated actress instead of exclusively representing real-life stars.

What's happening:

At the recent Zurich Summit, Van der Velden said studio attitudes shifted dramatically—companies that dismissed AI actors months ago are now interested in announcing which agency will represent Tilly "in the next few months"
Major studios are already piloting AI-assisted projects behind NDAs, according to Luma AI's Dream Lab LA head Verena Puhm, who noted some studios were "secretly working on stuff" while publicly denying AI use
Particle6 builds "hyperreal digital stars" from scratch, though technical details about the AI models powering Tilly remain limited

Addressing the backlash:

The rollout drew sharp criticism from performers' unions and actors concerned about job displacement. Van der Velden responded via Instagram, emphasizing they're "not trying to replace actors" but create a new type of genre, like CGI and animation.

Our take:

Calling it an 'AI actress' seems like a stretch - and is a label that's probably causing a lot of the controversy. This seems more like an 'AI influencer' - a character with a social following that can push brand deals. Artificial influencers aren't new - just look at Lil Miquela.

Blending Traditional VFX with AI

On X, Freddy Chávez Olmos shared a video demo of his tests combining Wan2.2-Animate with Nano Banana, Beeble, and Nuke for VFX work including enhancements, replacements, and full-head de-aging.

Experimenting with WAN 2.2 ANIMATE in combination with NANO BANANA, BEEBLE and NUKE for VFX enhancements, replacements and full-head de-aging.
— #Freddy Chávez Olmos (#@FreddyChavezO)
8:02 AM • Sep 26, 2025

Media Division tests the Lumix S1II with Arri LogC3 against the Alexa—comparing dynamic range, color, and workflow, with the S1RII on V-Log as reference.

Stories, projects, and links that caught our attention from around the web:

📸 Nikon patents a hybrid sensor that packs both global and rolling shutter modes in a single chip.

🌄 Tencent launches HunyuanImage 3.0, claiming the first open-source, industrial-grade multimodal model designed to deliver instant, high-quality visuals for creators.

🗣️ Wan just launched audio-driven video generation, letting users turn audio input directly into text-to-video and image-to-video clips.

Addy and Joey break down Alibaba’s Wan 2.5 — an AI video model rivaling Veo 3 with integrated audio and speech — alongside a wave of new tools from Google, Topaz, Nano Banana, and more in this week’s AI roundup.

Read the show notes or watch the full episode.

Watch/Listen & Subscribe

Spotify | Apple Podcast | YouTube