😺 ChatGPT-4o unleashed!

PLUS: What the 3rd biggest day in AI means for you

Welcome, humans.

The 1st biggest day in AI was on 11/30/2022 (ChatGPT’s debut). The 2nd was on 3/14/2023 (ChatGPT-4’s release). Yesterday was the 3rd, marking the launch of ChatGPT-4o.

Remember where you were when it happened, folks. We were personally trying to explain to our cat why it can’t sit on the keyboard during a Zoom call.

Here’s what you need to know about AI today:

ChatGPT-4o just dropped, and it’s free for everyone!
This new model has much stronger talking and seeing capabilities.
ChatGPT+ subscribers get 5x more access to GPT-4o than free users.
Looks like Google might announce something similar today!

On today’s podcast: Pete breaks down GPT-4o and hidden gems from the launch that people aren’t talking about enough (Apple Podcasts, Spotify, YouTube).

OpenAI unleashes ChatGPT-4o to everyone.

BREAKING: OpenAI announced, ChatGPT-4o (“o” for “omni”), the fastest, smartest, and most multimodal AI yet—watch the full demo here:

ChatGPT-4o will be available for free for everyone as a desktop app (soon). Yup, everyone will be able to use GPT-4o and GPTs as a result. This means you might want to rethink your Claude Pro or Gemini Ultra subscriptions.

ChatGPT+ users will get first dibs on GPT-4o with 5x more usage.

It’s more multimodal: This new ChatGPT isn’t just higher in IQ (it ranks #1 by far on the LMSYS leaderboard); it can talk and see just like us.

First, talking. We’ve had computers that can talk for a minute, but it never felt like a genuine convo—more a cycle of speak, pause, respond, pause, repeat.

Voice Mode feels like chatting with a real human—it captures your tone, language, and expressions in real-time. Many are describing it as a real-life Her (the voice in all the demos might actually be Scarlett Johansson).

Explore what it can do here:

live language translation (link).
realtime conversational speech (link).
lullabies and whispers (link).
sarcasm (link).
even singing (link)!

It’s uncannily human-like, perhaps too much so. But it means that for tasks you'd normally attempt with Siri, you should use ChatGPT instead. And with the new desktop app, Voice Mode will be great for scenarios better explained verbally than typed. Just remember to use inside voices in public spaces!

Here’s a quick guide on adding ChatGPT as a widget to your home screen, courtesy of Google SGE:

Second, ChatGPT-4o has live 20/20 vision, meaning it can interpret photos, screenshots, and docs while you work. For example:

Sal Khan’s son shares his iPad screen, and ChatGPT-4o helps him solve a problem live (link).
It can identify objects and teach you how to say them in Spanish (link).
It can explain copy & pasted code (link).
TOP DEMO: Be My Eyes + GPT-4o helping a blind person “see” what’s in front of him, even flagging down an available taxi (link).

Why it matters: Together, all these new features will unlock new use cases, and we’re pumped about them converging into a super helpful work assistant that can view your screen as you work.

Consider these possibilities:

Upload a PowerPoint and let ChatGPT-4o suggest layout tweaks, rephrase slide titles, and improve the design.
Use ChatGPT-4o to inspect a spreadsheet and highlight trends, anomalies, or discrepancies. Or for tech support.
GPT-4o can guide customers through visual step-by-step instructions for installing or setting up products.

Other updates not in the demo (see here):

For developers, GPT-4o is half the price, twice as fast as GPT-4-turbo, and has 5x rate limits.
Way better at writing text correctly in DALL-E 3 images.
It can create fonts.
It can generate 3D visualizations.

Catch Pete’s extended analysis on OpenAI’s demos and why the new desktop app is a game-changer (Apple Podcasts, Spotify, YouTube).

FROM OUR PARTNERS

Build AI apps in hours, not days.

Can you imagine it?

You…building custom AI applications that skyrocket your business…without worrying about being an AI genius? #promotion

With webAI’s Navigator, your business can build AI apps in hours, not weeks:

Speed up AI projects with models trained and fine-tuned in minutes and hours.
Maintain control over your data and its confidentiality.
Deploy seamlessly to local edge devices or cloud-based solutions.
Develop using full code or drag & drop, suitable for any skill level.

Be among the first to try Navigator by securing Early Access today!

Around the Horn.

Google's major developer conference, #GoogleIO, kicks off today at 1pm EST—we’re expecting some big AI announcements!
The three most common activities completed on ChatGPT include exploring new topics, researching products, and finding recipes.
Claude is now available for people and businesses across Europe.
Search engine volume will drop 25% by 2026, due to AI chatbots and other virtual agents, according to Gartner.