A Week of AI Drama (And What It Actually Means)

There's been a lot of drama in the AI world lately.

I don't usually talk about drama. But this time, the implications are real -- and worth understanding.

In the span of a few days, OpenAI released a major model upgrade and announced $110B in funding. ChatGPT uninstalls surged 295%. And Anthropic's Claude hit #1 on the App Store for the first time.

So today, I want to break down what’s been going on:

• OpenAI's new model and massive funding round

• What's driving the #UninstallChatGPT movement

• How Anthropic is turning a moment into momentum

And on a personal note, I did something a bit scary and nerve-wracking yesterday. I’m jumping back into “Founder Led Growth” for HubSpot. Details on LinkedIn (the post already has 14000+ likes and hundreds of comments).

Or to see the actual page itself, just go to GoHubSpot.ai.

Thanks for your support.

—@dharmesh

OpenAI's New Model + $110B Raise

Let’s start with the product news, which honestly got buried under the controversy.

OpenAI just released GPT-5.3 Instant, overhauling its default ChatGPT model for all users. The main differentiation is how the chatbot talks, not just how it thinks.

For months, users have complained about ChatGPT's tone -- overly cautious, preachy, what OpenAI itself called "cringe" in its release notes.

This new model:

• Dials back unnecessary refusals and tries to sound more natural.

• Has a reduction in hallucinations by over 25% on web search and nearly 20% on internal knowledge.

• Is a "stronger writing partner" with better web answers and information presentation.

Then came the funding news.

OpenAI announced $110B in new funding at a WHOPPING $730B valuation.

Yes, those are "Bs," as in billions. And remember, OpenAI is still private and approaching a trillion-dollar valuation.

But for me, the numbers, though big, aren't the most interesting part.

What's interesting is that OpenAI has forged a deeper partnership with Amazon, where OpenAI's models can now run on AWS infrastructure.

That's going to be welcome news for organizations that want to use OpenAI's frontier models like Codex -- but want to continue leveraging AWS's infrastructure.

This is a move that solves for OpenAI's enterprise customers (of which my company, HubSpot is one).

If I were to make a prediction, with $110B in the bank and AWS infrastructure access, OpenAI is positioning for massive enterprise expansion. They're going after the companies that wouldn't touch ChatGPT before because of data residency, compliance, or infrastructure requirements.

The AWS partnership removes those objections.

(NOTE: Although I am a small investor in OpenAI, none of the information in this post is non-public information.)

The #UninstallChatGPT Movement

But then the news that overshadowed everything: OpenAI's deal with the Department of Defense (rebranded under the Trump administration as the "Department of War").

Consumers across social media, particularly those outside the “AI Bubble,” started a trend to #UninstallChatGPT immediately after the news.

The backlash so far, has been significant:

• U.S. app uninstalls of ChatGPT's mobile app jumped 295% day-over-day on Saturday, right after Sam Altman posted the news (ChatGPT's typical day-over-day uninstall rate is 9%).

• Meanwhile, U.S. downloads for Anthropic's Claude jumped 37% on Friday and 51% on Saturday after Anthropic announced it would not partner with the U.S. defense department.

• ChatGPT’s app ratings also saw a surge in 1-star reviews: 775% on Saturday, then grew another 100% on Sunday.

This is real consumer sentiment, not just social media noise. This is the most significant user backlash I've seen to ChatGPT since its launch.

Curious, have you uninstalled ChatGPT and moved to Claude? Let me know in the poll below, and I’ll share the results in next week’s newsletter:

Anthropic's Strategic Response

Anthropic made a strategic marketing move in the midst of the whole drama: they made it easier to switch.

They launched a new tool called “Import Memory“ that lets users import their saved preferences and context from other AI providers with a single copy-paste.

Here's how it works:

1. Users copy a provided prompt into their current chatbot (ChatGPT, Gemini, or Copilot), paste the output into Claude's memory, and the switch kicks in within 24 hours.

2. The tool pulls saved instructions, personal details, project context, and behavioral preferences in a single upload.

Giving all those new users an easy way to bring their context over is a smart move for turning a viral moment into lasting retention.

My Take

Here's the thing: this isn't a simple story with a clear good guy and bad guy.

There's a legitimate moral question at the center. Should AI companies partner with defense departments? Different people will have different answers based on their values and experiences.

I'm not going to tell you which side is right. That's not my place.

BUT users voting with their app uninstalls is how markets are supposed to work. Companies responding to competition is also how markets are supposed to work.

In the end, the competition is making both products better.

Anthropic is building features that resonate because they have to compete. OpenAI is fixing issues faster because Claude is breathing down their neck.

Users win when companies have to fight for them.

What I’m paying attention to is: Will the uninstall trend continue? Will Anthropic convert these new users into retention? Will OpenAI's AWS push offset consumer losses? Will GPT-5.4 change the narrative?

I don't know. But I'm watching closely to see how this plays out.

—Dharmesh (@dharmesh)