- Neon Bees
- Posts
- OpenAI & Microsoft haggle
OpenAI & Microsoft haggle
& Perplexity's $8B valuation
Rise and Shine. A woman in Poulsbo, Washington, learned the hard way that feeding wildlife can get wildly out of hand. What started as a small family of raccoons decades ago turned into an invasion of 50 to 100 aggressive raccoons swarming her home. The furry mob became relentless, scratching at her doors, surrounding her car, and demanding food like she was running a 24/7 raccoon diner. Things got so bad that she had to call 911, reporting that the raccoons were acting aggressively and wouldn't leave her alone. Although feeding raccoons isn’t illegal in the state, authorities strongly advised her to stop. Thankfully, since cutting off their food supply, her unwanted visitors have finally started to move on, allowing her to reclaim her yard!
Top Stories
Microsoft, OpenAI Haggle Over Billion-Dollar Stakes
Justin Sullivan/Getty Images
Microsoft and OpenAI are deep in some old-school haggling, and the big question is: How much equity does Microsoft get for the $13.75 billion it’s pumped into OpenAI?
Since 2019, Microsoft’s been pouring cash into the AI powerhouse. Now, with OpenAI shifting from nonprofit to for-profit, it’s time to figure out how much of this AI empire Microsoft owns. Enter Goldman Sachs and Morgan Stanley—because when billions are on the table, you call in the big guns.
OpenAI’s value just skyrocketed to $157 billion after a funding round with heavy hitters like Nvidia, Thrive Capital, and SoftBank. Compare that to January 2023 when Microsoft’s $10 billion investment valued OpenAI at $86 billion. Not too shabby.
Despite all the hype, OpenAI’s still in the red, projecting a $5 billion loss this year. But they’re aiming for $11.6 billion in revenue next year. Ambitious? Absolutely.
With OpenAI restructuring into a for-profit public benefit company, Microsoft’s old profit-sharing deal is up for renegotiation. And CEO Sam Altman? He’s finally getting equity after years of having almost no skin in the game. Meanwhile, Microsoft’s not just negotiating for equity—they want more say in OpenAI’s governance too, after the surprise ousting of Altman left Satya Nadella in the dark. And when you’ve dropped billions, that’s understandable.
For now, it’s all about the numbers and boardroom drama.
Perplexity Eyes $8B Valuation Amid Growth and Challenges
Dado Runic/Reuters
Perplexity, the AI search startup, is aiming to double its valuation to a whopping $8 billion as it rides the wave of AI investment momentum. In just the past year, they’ve completed three funding rounds, with their valuation skyrocketing from $520 million in January to $3 billion by summer. Now, they’re looking to raise another $500 million, testing just how eager investors are for the next big AI unicorn.
Here’s what’s happening:
Rapid Growth: In under two years, Perplexity’s annual revenue has exploded—from $10 million in March to an impressive $50 million today.
Star-Studded Support: With heavyweights like Jeff Bezos backing them, Perplexity has some serious credibility as it pushes for even more growth.
AI Meets Search: Perplexity is like Google and ChatGPT rolled into one. It searches the web for real-time info, delivering instant answers. They’re monetizing through premium subscriptions and recently launched an enterprise search tool for businesses. Ads are coming soon, because why not diversify?
Content Drama: Not all’s been smooth sailing—Perplexity’s use of web content has ruffled some feathers. The New York Times even hit them with a cease-and-desist, raising concerns about their use of publisher material. CEO Aravind Srinivas says they’re working on smoothing things over, emphasizing that they’re not looking for beef with media outlets.
With 15 million queries a day and ads in the pipeline, Perplexity is on a fast track to becoming a major AI player. While they’re navigating some bumps along the way, their relentless growth and ambitious plans show they’re not slowing down anytime soon. Investors are watching closely—and so is everyone else.
Mira Murati Returns with New AI Venture and Big Ambitions
OpenAI
Mira Murati, former CTO of OpenAI and one of AI’s biggest movers and shakers, is diving back into the startup scene—and she's already making waves. While the details of her new venture are still top secret, she’s raising big bucks to build AI products powered by proprietary models. Word on the street is she’s gunning to raise over $100 million in funding. Why so much? Because, well, she’s Mira Murati. Her reputation alone has investors lining up.
But there’s more to the story. Barret Zoph, another ex-OpenAI heavyweight, left the company the same day as Murati. Coincidence? Probably not. Rumor has it that Murati’s been quietly recruiting other OpenAI talent for her new project, pulling together an AI dream team.
And let’s not forget her resume. Murati led groundbreaking projects at OpenAI, including ChatGPT and DALL-E, and played a key role in the game-changing Microsoft partnership. Her rapid rise in the AI world made her a household name, often sharing the spotlight with Sam Altman. Now, she’s following in the footsteps of other ex-OpenAI execs who left to launch their own ventures—think Anthropic and Safe Superintelligence.
The bottom line? With Murati at the helm, backed by top talent and a hefty stack of cash, her startup could be the next big thing in AI. Investors are clearly betting on it—and it’s safe to say the AI world is watching closely to see what she does next. Buckle up, because the Mira Murati show is just getting started!
X's New Privacy Policy Puts Your Tweets in AI's Hands
MEDIANAMA
X (formerly Twitter) just dropped an eyebrow-raising update to its Privacy Policy, and things are getting interesting. Starting November 15, third-party "collaborators" can use your X data to train their AI models—unless you say "no thanks" and opt out. So, if you’re not feeling like your tweets should help fuel the AI revolution, it’s time to do a little settings deep dive.
Elon Musk has already put X’s data to work training his xAI Grok chatbot, and now X is looking to make it a revenue stream, possibly licensing its data to anyone willing to pay. They’re following in the footsteps of Reddit and other media companies trying to cash in on their treasure troves of content.
The updated policy lays out that third parties could use your data for AI training and…well, pretty much anything else. The trick? Finding the opt-out switch. The policy points to the settings page, but doesn’t exactly give you a road map. Maybe it’ll get clearer when it goes live on November 15. Fingers crossed.
Another notable change? X can now hang onto your info as long as it feels like it—no more 18-month data expiration limit. And just because you delete a post doesn’t mean it’s gone for good. Search engines and other parties may keep their own copies around, because why make things easy?
Oh, and there’s a fun new “Liquidated Damages” clause. If you’re caught scraping more than a million posts in a day, X could hit you with a $15,000 fine per million scraped posts. So, yeah, no more mass scrapes unless you’ve got deep pockets.
All these tweaks come as X tries to monetize its data stash after losing advertisers and seeing subscription revenue fizzle out. Turns out, every post really does count when you’ve got bills to pay!
Gif of the day
Giphy
More Interesting Reads…
Insight of the day…
“Artificial intelligence is the science of making machines do things that would require intelligence if done by humans”