OpenAI: Out of Ideas
wq6e.comI saw an interesting quote from the deekseek CEO:
"When tackling a task, experienced people might reflexively tell you 'this is how it should be done,' while those without experience will experiment repeatedly, think deeply about the best approach, and ultimately find a solution tailored to the actual circumstances."
I do not think that OpenAI is out of ideas, but I do believe that large businesses have difficulty overcoming the innovator's dilemma. In addition to big businesses, there are also many small and very smart companies that are using investors' over exuberance to their advantage. Nevertheless, there are enough people trying that some are likely to find a way to develop higher-quality models at a lower cost.
> They’d rather give Sam Altman $200 per month for access to the world’s most mediocre researcher than get good results from a human.
Really going to need a citation for the claim that Deep Research is the world's "most mediocre researcher", the product was launched like 12 hours ago...
I'd buy it's not the best researcher in the world, but I'm willing to bet a lot that it's very far from the worst human researcher. I'd wager it's well above average.
I've felt this way since many of the other modalities they've announced to great fanfare fell by the wayside. Sora proved underwhelming. Likewise for voice mode. DALLE is far behind open alternatives.
It's hard to know because they're closed source, but evidence of advances in architecture is lacking.
OpenAI is not relevant in a world where DeepSeek is open source.
not sure if there would be a deepseek open source model if there were no leading models from OpenAI et al
just like Bell Labs UNIX gave way to BSD UNIX which gave way to SunOS/Solaris which gave way to Linux... this is a tale as old as time. The question is will this end the same way?
depending on whether the incumbents will keep the ability to ride new S-curves...
as in https://en.wikipedia.org/wiki/The_Innovator%27s_Dilemma
not sure if there would be an openai if it was not for the google paper in 2017 that introduced transformers.
The post complains about token cost and invisible tokens, but the feature is only available to subscribers who pay a flat rate (not per token).
While true, it is pretty telling that OpenAI feels like they need to resort to burning piles of cash. Generating a million output tokens in a month is trivially easy on the Plus plan which would set the company back 40$, per month, PER USER. Even in the whole blitz scaling model that we've apparently decided was only way to grow big bucks this is ludicrously unsustainable in the long run.
>but the feature is only available to subscribers who pay a flat rate (not per token).
???
Are you talking about API users? It's definitely available for them, though not all of them.
>o3-mini is rolling out in the Chat Completions API, Assistants API, and Batch API starting today to select developers in API usage tiers 3-5 (opens in a new window).
I think the question was if Deep Research is available from the API or not.
As far as I know it’s only available through ChatGPT and not the API. Same for o1 Pro Mode.
Anyone using ChatGPT is paying a flat rate and not paying per token. Thought they might run out of messages, based on what price they are paying.
The post is about Deep Research which I don't believe is an API feature.
Sure, right now it's only available to $200/mo subscribers. It won't stay like that forever...
This article is one of the dumbest takes I have ever seen. The author claims that OpenAI is out of ideas because they are only improving what they already have since GPT3.
That's like saying, getting a new computer is pointless, because it only has a better CPU, more memory and a better GPU, but is still the same parts, so it's pointless and they are out of ideas for making new computers.
One thing does not preclude the other. If someone keeps selling you incrementally better computers that consume more energy, and they also happen to sell you the energy you’re using, then (a) you should be suspicious and (b) you’d be right to suspect they’re out of ideas.
But you don’t necessarily need to stop buying the computers if it still makes sense to you.
My hope is on John Carmack....
My hot take is that these "reasoning" models (o3/R1, et. al.) are just silly stop-gap tricks. Any kind of agentic system built on foundational models is more than capable of outputting equal (or higher quality with tool usage) results. They are just shipping a highly generic version of that under the hood, and you're better off just custom tailoring an agentic process for your specific workflow rather than relying on theirs. There's a good reason Anthropic hasn't bothered coming out with one. I suspect these types of models will become a dead end as the foundational ones advance and agentic patterns become commonplace.
The author of the article is not an expert in this field. You should give him no more credence than you would give to, say, a plumber talking about astrophysics. Or astrpophysicist talking about plumbing.
The author of this comment is not an expert in the author of the article.
The author of the comment is, however, an expert in the field the author of the article is trying to (unsuccessfully) reason about.
If that were the case, surely you would have presented counterarguments instead of just attacking the author?
OpenAI has been releasing industry leading capabilities at a pretty predictable clip. Their current top reasoning model outperforms the competition by a mile. From this I conclude, at a minimum, that they aren't "out of ideas" and the article is just inane clickbait.
'The truth is they plateaued long ago with GPT-3'
Hmmm....
It wasn’t really until GPT-3.5 came out as ChatGPT that the world really started paying attention to OpenAI. I guess the author must have been one of the very few that had actually experimented with GPT-3 before the whole ChatGPT era.