Jay Neo left his role at MrBeast in 2023 after helping create some of the most viral content in YouTube history. Now he has launched Palo, an artificial intelligence platform aimed at helping creators analyze performance and generate fresh content ideas, according to TechCrunch.
The startup raised $3.8 million from Peak XV’s Surge program, with additional backing from NFX and individual investors.
Don't Miss:
- The AI Marketing Platform Backed by Insiders from Google, Meta, and Amazon — Invest at $0.85/Share
- Earn While You Scroll: The Deloitte-Ranked #1 Software Company Growing 32,481% Is Opening Its $0.50/Share Round to Accredited Investors.
How A Teen Hire At MrBeast Built One Of YouTube's Most Successful Formats
Neo joined the MrBeast team at age 18, initially focusing on viewer retention analysis. His work involved studying metrics to identify exactly where audiences lost interest in videos.
“I was so obsessed with retention graphs and figuring out why viewers stayed or why they left,” Neo told TechCrunch. “I had a document where I noted all this down.”
His breakthrough came with a simple concept: asking strangers on the street whether they would take a flight to Paris just to buy a baguette. That format generated more than 1.8 billion views across different channels, prompting MrBeast to produce multiple videos following the same structure.
After departing MrBeast, Neo co-founded several channels operating under the Creaky brand alongside another former MrBeast writer. Those channels reached over 1 billion views monthly, TechCrunch reported.
Trending: An EA Co-Founder Shapes This VC Backed Marketplace—Now You Can Invest in Gaming's Next Big Platform
Inside Palo's AI Engine For Hook Analysis, Viewer Signals, And Idea Flow
Palo requires creators to connect their social media accounts before the platform can begin analyzing their short-form video content. The system then evaluates what performs well and identifies underperforming elements.
Shivam Kumar, a former Palantir (NASDAQ:PLTR) engineer serving as chief technology officer at Palo, described the technical infrastructure behind the analysis. The platform employs multiple models to build a data structure capturing hooks, audience reactions, topic interest, originality, and potential search terms.
“The inference engine takes these primary data points and then uses a cocktail of top [large language models] to hierarchically aggregate these data points into cache for hot memory, embeddings which can later be semantically retrieved, and various other structured data formats,” Kumar told TechCrunch.
The system builds what Kumar called a "persona" for each creator based on their unique style and preferences. Creators can interact with the platform through a conversational interface similar to standard chatbots, TechCrunch reported. The tool can produce scripts based on proven formulas or create visual storyboards for creators who rely more on imagery than dialogue.
See Also: Missed Tesla? EnergyX Is Tackling the Next $200 Billion Opportunity — Lithium
Palo Opens Access After Large-Creator Trials Plus Strong Investor Support
Palo said it completed testing with approximately 40 creators who each maintained followings exceeding 1 million across various platforms. The company has now opened access to creators with at least 100,000 followers, according to its website.
The base subscription starts at $250 per month, with higher pricing tiers available for creators requiring expanded usage capabilities.
“Creators everywhere are looking for tools that make their process smoother without taking away their voice,” Peak XV Managing Director Rajan Anandan told TechCrunch. “Jay and the team had unusual clarity about where the real value lies and where it does not, which gave us strong conviction.”
Neo acknowledged concerns about AI pushing creators toward repetitive formulas. He compared the platform to how comedians refine material based on audience response.
“Here’s an analogy… when a comedian tries out some new material on the stage, they’re both consciously and subconsciously gathering data on whether the audience was amused or not,” Neo told TechCrunch. “Each performance becomes an iteration, and each new audience benefits from what the comedian learned from the show before.”
Read Next: Wall Street's $12B Real Estate Manager Is Opening Its Doors to Individual Investors — Without the Crowdfunding Middlemen
Image: Shutterstock
