😎Bezos' $6.2B AI Startup

PLUS: China’s New AI Billionaire | Perplexity Tops AI Failure Risk List

November 17, 2025

In partnership with

Reading time: 5 minutes

🗞️In this edition

  • Bezos returns to operations with AI engineering bet

  • China’s new AI power play

  • Perplexity named most likely to fail at SF summit 

  • Workflow Wednesday #45: Scaling Smarter

  • In other AI news –

    • OpenAI’s payments to Microsoft surge in leaked revenue docs

    • Anthropic says Chinese hackers used its AI to automate cyberattacks

    • Google commits $40B to build massive new data centers in Texas

    • 4 must-try AI tools

Hey there,

Jeff Bezos just dropped $6.2 billion on a new AI startup and took his first operational role since leaving Amazon. US export bans turned a nearly bankrupt Chinese chip startup into a $22.5 billion fortune for its founder in just six years.And Perplexity got voted "most likely to fail" by 300+ investors at a major SF conference despite its $20 billion valuation.

We're committed to keeping this the sharpest AI newsletter in your inbox. No fluff, no hype. Just the moves that'll matter when you look back six months from now.

Let's get into it.

What's happening:

Jeff Bezos is throwing his money and time into AI startup Project Prometheus, which he will help manage as co-CEO.

The company is coming out with $6.2B in funding, partly from Bezos, making it one of the most well-financed early-stage startups in the world, said three people familiar with the company.

This is the first time Bezos has taken a formal operational role in a company since stepping down as Amazon CEO in July 2021.

Project Prometheus is focusing on AI that will help in engineering and manufacturing in fields including computers, aerospace, and automobiles.

Bezos' co-founder and co-CEO is Vik Bajaj, a physicist and chemist who worked closely with Google's co-founder Sergey Brin at Google X. In 2015, Bajaj was among founders of Verily, a research lab dedicated to life sciences. Three years later, Bajaj co-founded and became CEO of Foresite Labs, an effort to incubate new AI and data science startups.

The company has already hired nearly 100 employees, including researchers poached from top AI companies such as OpenAI, DeepMind, and Meta.

Why this is important:

$6.2B funding makes Project Prometheus one of best-financed AI startups ever. For comparison, Thinking Machines Lab raised $2B, considered massive at the time.

Bajaj's background at Google X, Verily, and Foresite Labs brings deep experience in moonshot projects and life sciences AI.

Hiring 100 employees including researchers from OpenAI, DeepMind, and Meta shows aggressive talent acquisition competing directly with frontier labs.

Focus on AI for the physical world, not chatbots, positions Project Prometheus in a less crowded space than LLM development.

Our personal take on it at OpenTools:

Bezos launching a $6.2B AI startup as co-CEO is a massive bet that physical AI is the next frontier.

The timing is interesting. LLMs are mature. ChatGPT, Claude, Gemini all converging on similar capabilities. Physical AI, robotics, and scientific discovery are less developed.

$6.2B is a shocking amount for a startup in stealth. Bezos clearly funded most of it personally. His net worth is $230B. Spending $6B on an AI startup is ~2.5% of wealth. Significant but not crazy for someone his scale.

Physical AI learning from robots running experiments is a different approach than scaling LLMs. The question is whether it works. AlphaFold won the Nobel Prize, but that's one success story. Lots of physical AI bets have failed.

Bezos' aerospace interest aligning with AI for engineering and manufacturing makes sense. Blue Origin needs better manufacturing. AI that optimizes aerospace engineering directly benefits his other company.

This is Bezos' post-Amazon legacy play. Amazon is Jassy's company now. Blue Origin is space. Project Prometheus is AI. He's building a portfolio of moonshots.

From Our Partner:

Find your customers on Roku this Black Friday

As with any digital ad campaign, the important thing is to reach streaming audiences who will convert. To that end, Roku’s self-service Ads Manager stands ready with powerful segmentation and targeting options. After all, you know your customers, and we know our streaming audience.

Worried it’s too late to spin up new Black Friday creative? With Roku Ads Manager, you can easily import and augment existing creative assets from your social channels. We also have AI-assisted upscaling, so every ad is primed for CTV.

Once you’ve done this, then you can easily set up A/B tests to flight different creative variants and Black Friday offers. If you’re a Shopify brand, you can even run shoppable ads directly on-screen so viewers can purchase with just a click of their Roku remote.

Bonus: we’re gifting you $5K in ad credits when you spend your first $5K on Roku Ads Manager. Just sign up and use code GET5K. Terms apply.

What's happening:

Chen Tianshi's net worth hit $22.5 billion this year, more than doubling since January. His chip company Cambricon has surged 765% over 24 months.

The turnaround is dramatic. In 2019, Huawei cut off 95% of Cambricon's revenue to build its own chips. Chen's startup looked dead.

Then US export bans changed everything. Washington blocked Nvidia and AMD from selling high-performance AI chips to China. Beijing responded by mandating Chinese tech firms "buy local." Cambricon's revenue exploded 500% in the past 12 months.

Chen is now the third richest person in the world at or under 40, behind only Walmart and Red Bull heirs.

Interestingly, shares took off in August when Beijing urged companies to avoid Nvidia's H20 processors for government work. Cambricon warned investors its shares might be overvalued and that it still lags Nvidia by years.

Why this is important:

Chen's rise shows how US export controls created a protected market for Chinese chip makers.

State sponsorship, not product superiority, drove Cambricon's surge. "Buy local" mandates forced Chinese tech companies to source chips domestically whether they wanted to or not.

Questions about valuation are real. One analyst called the growth "mainly due to a low starting point" and warned valuations are "inflated without sustained policy support."

Our personal take on it at OpenTools:

This is what happens when geopolitics creates captive markets.

Cambricon went from near bankruptcy in 2019 to $22.5 billion founder wealth in 2025. That's not because their chips suddenly got better. It's because US export bans eliminated competition and Beijing mandated purchases.

The 765% stock surge reflects policy-driven demand, not technical breakthrough. Cambricon's own filings admit they lag Nvidia by years and face serious technical challenges.

Chen's academic pedigree is legit. PhD from elite Chinese university, groundbreaking papers on DianNao accelerators, early brain-inspired processors. But academic achievement doesn't explain 500% revenue growth in 12 months.

The parallel to DeepSeek is instructive. Both emerged from China's state-supported academic pipeline. Both benefited from export restrictions creating domestic demand. The difference is DeepSeek shipped a breakthrough model. 

Cambricon shipped chips because customers had no choice.

Nvidia's CUDA ecosystem is the real moat. Replicating hardware is hard. Replicating an entire software stack that developers already know is harder. Cambricon doesn't have that yet.

This wealth won't last if policies change or competition returns.

What's happening:

Over 300 founders, investors, and operators at the Cerebral Valley AI Conference in San Francisco voted Perplexity "most likely to fail" among high-valuation AI startups. OpenAI came second.

The poll, conducted during a closed-door session led by journalist Eric Newcomer, became the most discussed outcome of the event.

Perplexity hit a $20 billion valuation in September after raising $200 million. Despite positioning itself as a challenger to traditional search, investors say the company hasn't shown sustainable monetization or meaningful differentiation from larger AI players.

Amazon recently threatened legal action to prevent users from using Perplexity's Comet Assistant to make purchases on Amazon's platform. That adds regulatory risk on top of business model concerns.

Attendees described the vote as a "reality check" for startups relying on hype rather than proven revenue paths. Many warned that AI valuations have outpaced financial fundamentals, leaving aggressive scalers like Perplexity vulnerable to market correction.

Why this is important:

When your own industry votes you "most likely to fail," that's a credibility crisis.

The skepticism isn't about product quality. It's about business models. Perplexity raised $200 million at a $20 billion valuation without demonstrating how it makes money sustainably or defends against Google, OpenAI, and Microsoft.

Amazon's legal threat signals distribution risk. If platforms block Perplexity's assistant from e-commerce, that cuts off potential revenue streams before they start.

Our personal take on it at OpenTools:

This vote reflects what venture insiders see but won't say publicly.

A $20 billion valuation with no clear monetization path is bubble territory. Perplexity's competing against Google Search, OpenAI's ChatGPT, and Microsoft's Copilot without obvious moat. That's a tough position regardless of product quality.

The Amazon legal threat is a canary in a coal mine. If every major platform decides to block Perplexity's integrations, the product becomes less useful and distribution evaporates.

"Most likely to fail" followed by OpenAI is interesting. Even the industry's flagship company faces skepticism about sustainability at current valuation and burn rate.

The shift from "experimentation to financial accountability" is a key phrase. Fundraising got easy when AI was hot. Now investors want proof of revenue, not just engagement metrics.

Perplexity has strong brand visibility and a significant funding runway. But a brand doesn't save you if the business model doesn't work. The next 12 months will show whether they find sustainable revenue or become a cautionary tale.

This Week in Workflow Wednesday #45: Scaling Smarter – Growth-Focused AI Strategies

This week, we’ll show you how to use AI Tools for clear, actionable growth insights.

Workflow #1: Auto-Summarize Growth Analytics with Claude.ai
Step 1: Export your weekly metrics from Google Analytics, Beehiiv, or Stripe as a simple .csv or screenshot.
Step 2: Drop it straight into Claude and ask:

  1. Chaindesk - A no-code platform that allows users to create custom AI chatbots trained on their own data

  2. Iconik AI - A tool that helps users generate stunning app icons for Android, iOS, and web apps without any design skills

  3. Fill3d - A virtual staging tool that brings your empty room to life with photorealistic furniture

  4. Jason AI - A tool for automating B2B conversations and bookings

We're here to help you navigate AI without the hype.

What are we missing? What do you want to see more (or less) of? Hit reply and let us know. We read every message and respond to all of them.

– The OpenTools Team

How did we like this version?

Login or Subscribe to participate in polls.

Interested in featuring your services with us? Email us at [email protected]