Listen To The Show
Transcript
Welcome to The Prompt by Kuro House, your daily AI update. Today, we’ve got five stories packed with fresh product launches, legal moves, and breakthroughs that matter. Let’s dive right in.
First up, OpenAI has just launched a new app called Sora that lets you deepfake yourself for entertainment. According to Wired, Sora uses OpenAI’s latest video generation model, Sora 2, to create personalized AI-generated videos featuring your digital avatar. The app is currently iOS-only and invite-only, and it even adds AI-generated sounds to the videos for the first time in an OpenAI product. Users can control who can use their likeness, with options ranging from just themselves to mutual connections. OpenAI CEO Sam Altman highlighted safety guardrails built into the app to prevent misuse, including restrictions on sexual content, graphic violence, and extremist propaganda. It’s a fascinating blend of creativity and caution in AI deepfakes.
Next, a stealthy AI startup called Thinking Machines Lab has unveiled its first product, Tinker, which automates fine-tuning of frontier AI models. WIRED reports that Tinker is designed to make it easier for researchers and developers to customize powerful AI models like Meta’s Llama and Alibaba’s Qwen. Founded by former OpenAI leaders including Mira Murati, Tinker offers a user-friendly API that supports supervised and reinforcement learning fine-tuning. Beta testers say it strikes a rare balance between abstraction and control, letting users coax new capabilities from models without wrestling with complex infrastructure. The company plans to vet users to prevent misuse but aims to democratize access to frontier AI research. This could be a game-changer in how AI models evolve.
On the regulation front, California just passed SB 53, a first-in-the-nation AI safety and transparency law. TechCrunch explains that this bill requires large AI labs to disclose their safety protocols, especially how they prevent catastrophic risks like cyberattacks or bio-weapon creation. The law is enforced by the Office of Emergency Services and aims to hold companies accountable to their own safety promises. Adam Billen from Encode AI calls it a proof point that regulation and innovation don’t have to clash. While some industry players push for federal preemption of state laws, SB 53 shows that state-level action can be both practical and protective. It’s a significant step in balancing AI progress with public safety.
In transportation news, Waymo has been granted an extension to keep testing its robotaxis in New York City through the end of 2025. TechCrunch reports that Waymo can operate up to eight Jaguar I-Pace vehicles in Manhattan and Downtown Brooklyn, with human safety drivers behind the wheel. This permit exempts drivers from the usual rule requiring one hand on the wheel at all times. While this is a step forward, Waymo still needs separate licenses to carry passengers or run a commercial robotaxi service in NYC. The city currently lacks a permitting framework for fully driverless operation. So, the road to fully autonomous robotaxis in New York remains a work in progress.
Finally, looking at AI’s creative edge, TechCrunch highlights a panel at Disrupt 2025 where leaders from TwelveLabs, Wonder Dynamics, and Pocket Entertainment discuss AI’s role in storytelling and media. These pioneers are blending AI with human creativity to reshape animation, audio-first storytelling, and video understanding. The session promises insights into how AI is becoming a co-creator rather than just a tool. If you’re interested in where imagination meets machine intelligence, this is a must-watch event. Tickets are on sale now, with group discounts through October 3rd.
That’s a wrap for today’s AI update on The Prompt. We’re seeing exciting advances in AI tools, thoughtful regulation, and the ever-expanding horizons of creativity. Thanks for listening, and we’ll catch you tomorrow with more news that shapes the future.
