OpenAI dropped the price of o3 by 80%

I am not saying they haven't improved the laziness problem, but it does happen anecdotally. I even got similar sort of "lazy" responses for something I am building with gemini-2.5-flash.

This is marginally less true for embedding models and things you've fine-tuned, but only marginally.

Example, I asked it to write something. And then I asked it to give me that blob of text in markdown format. So everything it needed was already in the conversation. That took a whole minute of doing web searches and what not.

I actually dislike using o3 for this reason. I keep the default to 4o. But sometimes I forget to switch back and it goes off boiling the oceans to answer a simple question. It's a bit too trigger happy with that. In general all this version and model soup is impossible to figure out for non technical users. And I noticed 4o is now sometimes starting to do the same. I guess, too many users never use the model drop down.

OpenAI is very good at this as well because of their brand name. For many people ChatGPT is all they know. That's the one that's in the news. That's the one everybody keeps talking about. They have many millions of paying users at this point.

This is a non trivial moat. If you can only be successful by not serving most of the market for cost reasons, then you can't be successful. It's how Google has been able to guard its search empire for a quarter century. It's easy to match what they do algorithmically. But then growing from a niche search engine that has maybe a few tens of thousands of users (e.g. Kagi) to Google scale serving essentially most of this planet (minus some fire walled countries like Russia and China), is a bit of a journey.

So Google rolling out search integration is a big deal. It means they are readying themselves for that scale and will have billions of users exposed to this soon.

> Their last funding round in March valued them at 300B. Despite losing 5B last year, they are growing really fast

Yes, they are valued based on world+dog needing agentic AIs and subscribing to the extent of tens or hundreds of dollars/month. It's going to outstrip revenue things like MS Office in its prime.

5B loss is peanuts compared to that. If they weren't burning that, their ambition level would be too low.

Uber now has a substantial portion of the month. They have about 3-4 billion revenue per month. A lot of cost obviously. But they managed 10B profit last year. And they are not done growing yet. They were overvalued at some point and then they crashed, but they are still there and it's a pretty healthy business at this point and that reflects in their stock price. It's basically valued higher now than at the time of the Softbank investment pre-IPO. Of course a lot of stuff needed to be sorted out for that to happen.