Happy Wednesday.
The current thing in tech and business is the mixed reaction to Google’s announcements on the first day of its annual developer conference I/O.
Today’s lineupFigma Co-Founder & CEO Dylan Field at 11:30 AM
Socket Founder & CEO Feross Aboukhadijeh at 11:50 AM
Key Context Founder Tae Kim at 12:00 PM
Mercury Founder & CEO Immad Akhund at 12:30 PM
Airbnb Co-Founder & CEO Brian Chesky at 12:45 PM
Minted New York Founder Marcus Milione at 1:10 PM
Reactions to Google I/O: Day 1
Yesterday was the first day of Google’s annual developer conference I/O, which is being held in Mountain View. The company’s leadership team announced a number of new initiatives, products, and upgrades to core features of their business. The reaction online has been mixed.
Intelligent Eyewear
In a partnership with Samsung, Warby Parker, and Gentle Monster, Google is developing two pairs of smart glasses — one audio-only, and the other with a display. The audio glasses are set to launch this fall, and will feature speakers at your temples and cameras on the frame, allowing you to ask questions about what you’re seeing. The built-in display glasses are still in development, but Google says they’ll be able to show you the standard set of stuff: texts, images, directions, etc.
Gemini Omni
Yesterday Google demoed its new AI video generator, which is Sora-like in a number of ways (namely, you can upload selfies and use the tool to put yourself in fun situations, environments, etc). Demis described Omni’s features on 𝕏: “It can take photos, video & audio and build entirely new scenes. Over time it’ll be able to handle any input & any output - starting w/ video”; watch the launch video here.
Upgrades to Gemini LLMs
The company announced two AI model updates yesterday: Gemini 3.5 and the more affordable Gemini 3.5 Flash. Developer reaction was rough; suffice to say, 3.5 is decidedly not a GPT 5.5 or Opus 4.6 moment.
Antigravity
Google’s agentic dev platform Antigravity, which originally launched in November 2025, got a massive overhaul. The company added a CLI and SDK — features the tool didn’t previously have — and a standalone desktop app. Here’s the launch video.
Antigravity has a very interesting backstory. The product was created by Varun Mohan and Douglas Chen at a company first named Codeium and then renamed Windsurf. It was a direct competitor to Cursor, and was seeing solid traction when Google came in with a $2.4 billion licensing agreement and employment agreements for Varun, the CEO, and many of the engineering team members. It was the highest profile “zombie acquisition” of last year, with a sort of “ghost ship” shell left behind, which was quickly picked up by Scott Wu, the CEO of Cognition. So today, Windsurf is a Cognition product, that sits alongside their AI software engineer Devin, many of the Windsurf team members work at Cognition, and Google sort of rebuilt the original Windsurf product into what is now Antigravity.
Reviews are still flowing in for the product, but what’s interesting is the go-to-market motion that will be required to really get it growing. Even though Google Cloud is massive (and notably growing faster than AWS and Azure right now), it’s unclear if distribution will flow top-down through enterprises (Microsoft has GitHub through which they can push Copilot), or bottom-up through Gemini App users who graduate to vibe coding and stay on the Google AI bundle. No matter how they solve it, there’s immense value to being a front-door AI capabilities, whether that’s through a chat interface, an agent, or an IDE.
— John and Brandon
Clip Spotlight: Shazam Founder Philip Inghelbrecht on the early, early days of ShazamNot many people remember this, but the first version of Shazam in 2002 required you to dial a four-digit number, hold your phone up to a song, and wait for a text message back with the track name and artist.
Even then, the company was already running a music recognition algorithm to identify songs, years before the iPhone or iTunes Store existed.
Shazam founder Philip Inghelbrecht explains:
“We put [Shazam] together before the iPhone or the iTunes Store existed. So with the first version, which launched in August of 2002, when you heard a song, you actually had to take your phone and dial a short code. 2-5-8-0. You didn’t have to remember the number because if you look on any telephone, 2580 are the four digits right in the middle.”
“We then would listen to the song as if you were speaking into your headset, do the recognition, and then send a text back with the name of the track and the artist.”
“The algorithm was real, but the experience was just clunky. It wasn’t until the iPhone came along, where you had that beautiful experience with a touch color screen and it comes back with rich information, and that changed everything. Not to mention the distribution with the iTunes Store.”
HeadlinesUnherd story on TBPN: What tech bros learned from ESPN True fans always want more
Google partners with Warby Parker and Gentle Monster on intelligent eyewear
NYT: Meta Lays Off 8,000 Employees, As A.I. Casualties Mount
Reuters: Exclusive: Intuit to cut 17% of global jobs to streamline operations, memo shows
Bloomberg: Goldman Sachs to Lead SpaceX’s Mega-IPO Bank Lineup
The Information: SpaceX IPO Could Return More Than $60 Billion Each for Founders Fund, Valor
WSJ: See How SpaceX Is About to Eclipse Every Other Blockbuster IPO
CNBC: Jeff Bezos says bottom half of earners should pay zero in income taxes
NYT: James Murdoch, Intent on ‘Thoughtful Journalism,’ Buys Half of Vox Media
The Information: Sam Altman Offers YC Founders $2 Million in OpenAI Tokens For Equity
FT: Nvidia’s Jensen Huang bankrolls AI boom with $90bn deal spree
WSJ: How Zyn Became All the Rage Inside Trump World—Including With RFK Jr.
Mercury raises $200M Series D at a $5.2B valuation led by TCV
Nate Silver: Disney erased FiveThirtyEight
Nan Ransohoff: The third wave of American philanthropy
WSJ: CEO Walks Back Comment About Replacing ‘Lower-Value Human Capital’ With AI
Posts of the Day