PRODUCTS

Google's Project Genie generates interactive worlds

Google is widening the world of world models. 

On Thursday, the company’s AI unit released Project Genie, an “experimental research prototype” that lets users create and explore virtual worlds and video-game like environments. Currently available to Google AI Ultra subscribers in the US, the system is powered by Gemini, Nano Banana Pro and Genie 3, its most powerful world model yet, released in August. 

Project Genie offers a number of features for users to play around with, including:

• World sketching, in which users can prompt the model with text and images to create a “living, expanding environment;” 

• World exploration, allowing users to navigate these developed worlds by generating paths ahead in real time based on user actions; 

• And world remixing, letting users build on top of existing prompts to edit and curate the world to their liking. 

Google admitted that there are still some bumps to be ironed out with Project Genie, noting that generated worlds might not be completely true to life or adhere entirely to real-world physics. Generations are also limited to 60 seconds, and characters aren’t always entirely controllable.  

However, the experimental model is the latest step in its broader journey to “support [it’s] AGI mission,” the company said. “Building AGI requires systems that navigate the diversity of the real world.” 

Google’s release comes as world models receive more and more attention. Last week, Bloomberg reported that World Labs, a world model startup founded by Fei-Fei Li, Ph.D., is in talks to raise hundreds of millions at a valuation of $5 billion. AMI Labs, a world model firm founded by Yann LeCun, is also rumored to be discussing a funding round that would notch the company a $3.5 billion valuation. 

Capable of reflecting the real world physics and consequences, many are eyeing world models as the key to achieving AGI, rather than simply scaling large language models to as many parameters as possible.

Google has a number of advantages in the AI race. Its cloud prowess, internet legacy, and healthy well of cash have given it a solid foundation, allowing it to both build powerful large language models and experiment with newer technologies. Unlike other AI firms that need to primarily focus on building up their foundational models as a means of making money, Google has the leeway to branch out and try new things, whether that manifest as world models like Project Genie, physical AI like Gemini Robotics or image understanding like Agentic Vision. Along with enabling it to reach new product markets, this broad experimentation could also inch the company closer to the elusive goal of AGI.

TOGETHER WITH DECAGON

Your Customer’s Voice Matters More Than Ever

Whether via surveys and support conversation or product reviews and NPS scores, customer feedback has always been useful for brands –  but very few are maximizing its value. That’s because feedback often comes after the fact, doesn’t get total visibility by your team, or lacks actionable insights and next steps. 

The good news is, these are all problems that AI can help solve. In their latest ebook, “Why Voice Of Customer Matters More Than Ever”, Decagon explores how AI is turning customer convos into real-time intelligence for your brand, the ways it can empower your product, CX, and engineering teams with consumer feedback, and why it can even turn support from a cost center into a growth driver. 

Get your free copy of the ebook right here (and say thanks to Decagon!).

CONSUMER

OpenAI social platform may trade data for bot control

The Twitter-turned-X feed is riddled with bots. OpenAI’s network seeks to remedy the issue — but at what cost?

The AI company is working on a humans-first platform positioned to compete against dominant social media firms, such as X, Instagram and TikTok, according to a new Forbes report. The app is still in the early stages, with a small team of less than ten people developing it. 

The key differentiator between this platform and others is that  it may include a biometric identity component. Sources familiar with the project cited in the report said the team is considering requiring users to prove their “personhood” via Apple’s Face ID or the World Orb, an eyeball scanner developed by Tools for Humanity, a company co-founded by Sam Altman. 

World’s Orb is a sphere that includes an optical system with telephoto lens and 2D gimbal mirror, created for the sole purpose of proving humanity, according to the site. The company claims the iris image is encoded and encrypted directly to the World Orb, keeping user data safe. Naturally, concerns persist about both the privacy risks of handing over biometric data and whether it is appropriate to mandate this at all.

This report breaks as new data shows that OpenAI’s first tiptoe into the social media space, its Sora app, is declining in popularity, as first reported by TechCrunch. 

The Sora app, which let people share AI video and image generations, became a viral sensation upon launch in September, accruing 100,000 installs on day one despite being only available on iOS and an invite-only experience. It then climbed to one million downloads, beating a record ChatGPT set as the fastest growing app of all time. However, the app has since lost momentum as Appfigures data shows downloads fell 32% and 45% month-over-month in December and January, respectively.

OpenAI has long shown interest in launching a new social media platform, with the first reports circulating as early as April 2025. However, it doesn’t seem like OpenAI is operating out of a desire to simply enter the social media space. Rather, the company is trying to rewrite the rules of social media as we currently know them. For example, with the Sora app launch, it released a new kind of customizable recommender algorithm that could be tweaked through natural language, putting control of algorithms back in users’ hands. With this new project, it is also challenging the status quo, attempting to get rid of bots and recenter humanity. It may also be an attempt to capture some of the digital ad revenue that social media platforms dominate.

TOGETHER WITH DECAGON

How To Turn Customer Feedback Into Big Wins

Despite what some people might think, feedback is a very, very powerful thing… if you know what to do with it. The problems occur when your business uses that feedback incorrectly – getting it too late, not having proper visibility into it, or simply having no clue how to turn it into actionable insights. 

And that’s exactly where this new ebook from Decagon can help. Inside are real tactics on how to improve your customer feedback methods, how to turn conversations into structured insights, and even a framework for transforming support from a cost center into a growth driver. 

Download your free copy right here and start making the most of your customer feedback. 

CULTURE

Music industry remains split on AI

The music industry’s back-and-forth battle with AI wages on. 

On Wednesday, a coalition of music publishers including Universal, Concord and ABKCO filed a copyright infringement lawsuit against Anthropic, seeking more than $3 billion in damages over the infringement of more than 20,000 songs. According to a report from Music Business Worldwide, The complaint intensifies a lawsuit filed in October 2023, which covered roughly 500 songs and sought $75 million in damages. 

The plaintiffs told Music Business Worldwide that the lawsuit is the result of the “persistent and brazen infringement of our songwriters’ copyrighted compositions taken from notorious pirate sites.”

Along with targeting AI companies, music platforms are now targeting the AI-generated music itself: On Thursday, Deezer made its AI music detection tool commercially available to other platforms. The tool finds, tags and removes AI-generated music from its recommendation algorithm with 99.8% accuracy, according to the release. In 2025, Deezer’s tool caught more than 13.4 million tracks created using AI. 

But the music industry may be sending AI firms mixed signals. Late last year, Warner Music Group settled its lawsuits with AI platforms Suno and Udio. Universal Music Group, meanwhile, settled its own lawsuit with Udio, announcing a partnership with the startup, and forged a deal with Stability AI. 

Musicians themselves are divided on where AI belongs in their creative processes. In November, hundreds of artists, including Paul McCartney, Kate Bush and Annie Lennox, came together to create a silent AI protest album titled “Is This What We Want?” Others are opting to go with it, with artists like Liza Minnelli and Art Garfunkel partnering with ElevenLabs on a project to “co-create” studio-quality music.

What this fight boils down to is control. Record labels and publishers want control of the musical outputs that these AI models are creating, as well as any potential cash that can be made as a result. This is likely why these labels are striking deals with smaller AI firms while legally going after the big ones that can’t be so easily swayed. Artists, too, want control over these outputs, especially given the hard work and dedication to the craft that goes into writing music in the first place. Creation is the core of the music industry. Without control over that creation, the potential to make a living as a creator is existentially threatened.

LINKS

• OpenAI reportedly prepares for IPO in fourth quarter of 2026

• Nvidia, Microsoft, Amazon to join OpenAI’s twelve-figure funding round

• ByteDance, Alibaba set to launch new models 

• OpenAI’s translation tool now supports 47 languages “and counting”

• Perplexity, Microsoft ink $750 million, three-year cloud deal 

• AI-assisted coding causes “statistically significant decrease” in mastery 

• Waymo robotaxi struck a child near elementary school in Santa Monica

• Tesla to end production of Model S and Model X

• Airtable Superagent: The first standalone product from Airtable, a bet on “multi-agent coordination” that deploys several agents to tasks in parallel. 

• Trinity: A 400-billion parameter open source foundational model by Acree AI, one of the largest of its kind by a US company.  

• Mureka V8: A new audio AI model using “Music Chain of Thought” with improved human-like logic and valid musicality. 

• Willow Voice: A voice-to-text typing experience available on mac, windows and iOS.

The AI Talent Bottleneck Ends Here

U.S.-based PhD researchers can’t keep pace with the demand for evaluation and reasoning work.

Athyna Intelligence connects AI teams with vetted LATAM PhDs and Masters in U.S.-aligned time zones.

Same expertise, faster access, 40–60% savings.

Meet PhD experts

(sponsored)

GAMES

A QUICK POLL BEFORE YOU GO