“There is no innovation and creativity without failure. Period.”

—Brene Brown, Author

OpenAI launched Sora 2 in late September 2025 with a standalone app and positioned it as its flagship video-and-audio generation model. It climbed to #1 in Apple’s App Store, and even lead to signing a $1B three-year Disney deal covering more than 200 licensed characters.

And then set a shutdown timeline. That’s a busy few months.

Sora came in hot, then watched traffic drift toward Runway, HeyGen, Kling, and the rest of the field.

One line in OpenAI’s own help docs says a lot: “look forward to Sora for Business.” That points the product story away from viral clips toward a more controlled environment with clearer economics, tighter permissions, and fewer moderation headaches.

Sora web and app experiences will end on April 26, 2026, and the API will end on September 24, 2026.

Dealerships already rely heavily on video, so it’s important to build around workflows your team controls.

Keep the raw footage. Store templates locally. Save brand assets outside the platform. Use AI tools that connect to inventory, CRM, and ad operations, where the output ties back to appointments, engagement, and sales activity.

AI tools can drift with the current. Your internal systems need to anchor the boat.

Google’s new Lyria 3 rollout gives AI music a more practical job.

Lyria 3 Clip creates 30-second tracks for social posts, loops, and quick testing, while Lyria 3 Pro makes longer songs (up to three minutes) and gives users more control over structure, including intros, verses, choruses, and bridges.

Google says the model also handles vocals, lyrics, tempo, mood, and even image-based prompting with more control than earlier versions.

That matters because most AI music tools still feel like a sketchpad. Lyria 3 feels closer to working gear.

Google is putting it inside the Gemini API, Google AI Studio, Vertex AI, Google Vids, the Gemini app, and ProducerAI, which means teams can use the same model across content creation, product workflows, and internal video production.

For dealership teams, that opens up easy use cases: music beds for inventory walk-arounds, service promos, recruiting videos, event recaps, and podcast intros.

Once audio becomes this easy to shape, more teams will stop reaching for stock tracks and start building something that actually sounds like them.

Google has a great guide for prompting with Lyria here, so give it a whirl and let us know what you come up with!

Most teams already have AI running inside the store. The real question is how precisely it’s being used.

In this episode of The Dealer Playbook, Kristine Lentz walks through how real-time OEM data reveals exactly when a customer leaves your pipeline and what that timing says about your process. Those signals can guide where AI supports follow-up, where it extends engagement, and where it hands off to a more intentional service-driven approach.

Watch the full episode to see how this level of precision changes the way your team works.

(Looking for even more expert insight? Check out this year’s brand new AutoIndustry.Ai Summit.)

Thanks for reading, Friend! For our money, we’ll still take authenticity over artificiality any day of the week.