... And maybe something like this is, it was probably just too early.
9,073 karma · joined May 10, 2020
... And maybe something like this is, it was probably just too early.
Slight nuance - they have spent a [reasonable amount of money] automating production.
The trick to automating something that ‘isn’t a car’ is often to put in small bits of low-cost and flexible automation that can be moved around and repurposed. IMO this is often what we are bad at in the west - companies can/do setup massive automated sites at huge expense, but there aren’t the skills/infrastructure to do this at the lower end of production (eg if you want to deploy one AMR in the west the AMR companies don’t want to talk to you, and there isn’t really an easy way to get one yourself without talking to an integrator which will charge tens of thousands which will wipe out the benefit, and we don’t have the skills within most small production companies to get a small robot arm/AMR working without external integrators - but a one-AMR deployment might be a more common scenario in China).
> Durham University improved by 30 places year-on-year in its students’ evaluation of teaching quality, which was the main driver in securing its third place in the overall university league table
Which isn't quite the same as 30 places in ranking as OP suggests, however I agree with their point that moving 30 places on that metric could be fairly suspicious.
For example - when I was at university in the UK we got a speech telling us basically that we were going to get sent a survey from the times, and the higher we ranked the university, the higher the universities ranking would be, and that would make our degree more valuable. If the main reason they jumped from 7th to 3rd could be a metric that is potentially 'influence-able' by the university, it could be more of a change in comms-strategy than actual university quality.
The 'few months' bit doesn't seem quite right - the cost to get a human to drive a car would maybe be $50k per year, so i'm not sure how a $200k vehicle can pay itself off in a few months vs a $50k car + $50k per year driver.
I'm aware that the cost / ride is higher for Waymo, but it doesn't sound like that would be enough to cover the extra $200k and not certain that scales to other geographies outside of SF.
I mean to pay off a $200k vehicle in a few months you would need each car to be clearing $3k a day in revenue or something like that.
It's probably a few years per car if they are at $200k. If they are 'in the black' or not will probably depend more about their accounting rules (i.e. depreciation) more than anything else.
As a counterpoint, if OpenAI were actually profitable at this early stage that could be a bad financial decision - it might mean that they aren't investing enough in what is an incredibly fierce and capital-intensive market.
At the research level it’s not just about being smart enough, or being a good programmer, or even completely understanding the field - it’s also about having an intuitive understanding of the field where you can self pursue research directions that are novel enough and yield results. Hard to prove that without having done it before.
I think a lot of them relied on gradual improvement and lots of 'mini-breakthroughs' rather than one single breakthrough that changes everything. These mini-breakthroughs took decades to realise themselves properly in almost every example on the list too, not just a couple of years.
My personal gut feel is that even if the core technology plateau's, there's still lots of iterative improvement to go after on the productisation/commercialisation of the existing technology (e.g. improving tooling/ui/applying it to solving real problems/productising current research etc).
In electric car terms - we are still at the stage where Tesla is shoving batteries in a lotus elise, rather than releasing the model 3. We might have the lithium polymer batteries, but there's still lots of work to do to pull it into the final product.
(Having said this - I don't think the technology has plateau'd - I think we are just looking at it across a very narrow time span. If in 1979 you said that computers had plateau'd in 1979 because there hadn't been much progress in the last 12 months they would have been very wrong - breakthrough's sometimes take longer as technology matures, but that doesn't mean that the technology two decades from now won't be substantially different.
I think the point of most (frontier) small models is usually to provide the best answer possible given small inference resources, rather than to reduce training time.
This is more of a toy model, so fun and an interesting project but it doesn't necessarily tell us what the art of the possible is for small models.
Sure they still have massive problems with hallucination, but this article doesn’t give us any more insight into that I don’t think!
Go open the OpenAI API playground and give GPT3 and GPT5 the same prompt to make a reasonably basic game in JavaScript to your specification and watch GPT 3 struggle and GPT 5 one-shot it.
By finding and testing problems that o3 can't do on Deep Think, and also testing the reverse? Or by large benchmarks comparing a whole suite of questions with known answers.
Problems that both get correct will be easy to find and don't say much about comparative performance. That's why some of the benchmarks listed in the article (e.g. Humanity's Last Exam / AIME 2025) are potentially more insightful than one person's report on testing one question (which they don't provide) where both models replied with the same answer.
It would be more interesting to know if it can handle problems that o3 can't do, or if it is 'correct' more often than o3 pro on these sort of problems.
i.e. if o3 is correct 90% of the time, but deep mind is correct 91% of the time on challenging organisational problems, it will be worth paying $250 for an extra 1% certainty (assuming the problem is high-value / high-risk enough).
I’m saying if the goal is to invent true robotaxis (because that’s the market you are going after) then I think waymo’s approach of following what works and cost-optimising later gets you there faster if you have enough funding.
Tesla on the other hand started with a load of constraints - smaller amounts of compute, cheaper sensors, needs to hit a price point as it will be installed in every vehicle sold - can't be $50,000 of dedicated self driving hardware per vehicle.
There are some scenarios where constraints lead to breakthroughs, but often for true-moonshot projects it is the opposite (see: Space shuttle, System/360, the manhattan project, LHC etc)
i.e. if I open two tabs, and edit one tab, the other tab doesn't update and visa-versa until refresh.
You could probably hammer the most expensive cursor API all-day every-day and it would still be a fraction of the cost of a junior dev.
IMO don't think this is ready for production use in the slightest - but cool project!
* Can create multiple workspaces with same name which then ticks both * Invite user seems to not work randomly or will not send the email * Cards with special characters like @ will just not be created, and won't show error messages.
The formatting looks all off for me on chrome mac (black bars) and then if I click on a card it opens a window but then doesn't load any data.
Also it looks like a bug that if you filter by some tags, then click a card, the filter gets reset.
This is 5 secs of testing on the one board you have publicly shared, so there might be a few bugs to iron out!
My experience is the opposite - I'm using it in Cursor and IMO it's performing better than Gemini 2.5 Pro at being able to write code which will run first time (which it wasn't before) and seems to be able to complete much larger tasks. It is even running test cases itself without being prompted, which is novel!
But personally I can't see how the popularity of oat and almond milk in indepenent coffee shops tells us that much about how people perceive the inner workings of chatGPT.
Of course it's an indicator of future performance - Not a guarantee, but certainly a indictator.
People usually see the product rather than the process.
Extrapolating that technology will get better in the future when it has got better in the past isn’t a sure bet, but it’s a reasonably reliable one.
In 2 years we have moved from AI video being mostly a pipe dream to some incredible clips! It’s not what this is like now, but what will it be like in 10 years!
Also the term 'LLM' is more about the mechanics of the thing than what the user gets. LLM is the technology, but some sort of automated artificial intelligence is what people are generally buying.
As an example, when people use ChatGPT and get an image back, most don't think "oh, so the LLM called out to a diffusion API?" - they just think "oh chat GPT can give me an image if I give it a prompt".
Although again, the term is entirely abused to the extent that washing machines can contain 'AI'. Although just because a term is abused doesn't necessarily mean it's not useful - everything had "Cloud" in it 10 years ago but that term was still useful enough to stick around.
Perhaps there is an issue that AI can mean lots of things, but I don't know yet of another term that encapsulates the last 5 years advancements in automated intelligence, and what that technology is likely to be moving forwards, which people will readily recognise. Perhaps we need a new word, but AI has stuck and there isn't a good alternative yet, so is probably here to stay for a bit!