Look, Another AI Browser

To find out what someone truly believes, don't listen to what they say, observe how they act. I don't see how OpenAI's recent actions make any sense from the perspective a company that internally believes it's actually close to unlocking super-intelligence.

OpenAI has always had the stance of “commercialize narrow AI that is research aligned with AGI development”. In fact they used to ask this as an interview question — “should we commercialize narrow AI or aim to put all resources into AGI”. The correct answer required you to prove you drank the kool aid and also wanted to make tons of money.

I think they need to respond to all the funds they've raised and need to generate money somehow beyond subscriptions.

No, but their actions do suggest they think they're nearing a disruption to both browser and web page: a new way to acquire and make use of information.

Like an information OS for the information cloud.

Is there a correct answer?

Makes sense, I guess. Did you hear that question yourself in an interview, hear it from someone who interviewed, or hear that as a story through the grapevine? and ~when was it asked?

They are multiple companies in-one. One that is pushing for AGI, model development, one that is trying to build consumer apps and "win" AI applications/platform moat.

I suspect that this will always be just one more year away like Tesla's robotaxis

Commercializing something gives you more resources to do the thing you really want to do.

Anthropic came to the same conclusion. SSI either doesn't have something commercializable or came to a different conclusion.

I'd go a step further: has OpenAI actually achieved any significant research breakthroughs on par with Google's transformers? So why does everybody think they will achieve the next N breakthroughs necessary to get to AGI?

They basically invented LLMs as we know them (autoregressive transformers trained on web data) with the GPT 1/2/3 series of papers.

They also pioneered reasoning models, which are probably the biggest breakthrough since GPT-3 on the AGI tech tree.

They're spinning up their own advertising platform; chatgpt is a coherent contender to google's search bar, over a long enough time span if they can maintain user engagement numbers, it seems plausible that they could secure half of google's ad revenue. Spinning up a browser is not cheap but it certainly lines up with their actions of spinning up a perpetual advertising machine via chatgpt to fund other things. Which might include AGI if/when that happens.

>a company that internally believes it's actually close to unlocking super-intelligence

I am not sure if this is still true, they started backing off from this line of reasoning summer of 2024 and haven't returned to it.

Google invented transformers. OpenAI just released their model to the public first. Good for them, but not exactly impressive research.

Reasoning models are pretty cool, but it was just taking what every body was already doing manually ("and please show your work") and making it automatic. The whole agentic shift is also nice but kinda obvious. But I'm still struggling with hallucinations and context rot all the time and it's becoming increasingly clear that that's not something that can be solved incrementally. We need more architectural breakthroughs like the transformer to achieve something like real AGI. Possibly several more.