This is a pretty moronic take. You only have to understand the first thing about physics (e.g., a 14-year-old's understanding) to know that a dragless airplane is impossible.
If you are suggesting there is some new model of physics or groundbreaking technology that would allow such a dragless plane, then don’t let me discourage you! But AI won’t help at all, since it will only regurgitate conventional wisdom…
> I wouldn't be surprised if there are now more chess books now than there ever have been
Well yeah... how would there be fewer??
But the point itself is silly. Few people are putting effort into Maths for the fun of it (and of those that are many derive fun from being the only one who can produce a solution). Chess differs in that it never had any point but the game its self.
There’s always been a distinction between published work and unpublished.
If my friend showed me some unpublished work (Say 90% of the hardest work toward a significant proof) if I then go finish the remaining 10% and publish the whole thing as mine it isn’t ‘building on the shoulders of giants’ it’s plagiarism/theft.
If he’d published that work first and I took it and found another proof and credited him for his work via citation then that would be fine.
It’s more like you spend 4 years developing a product you’re passionate about. This product will gain you the respect of all your colleagues and either earn you money directly or lead to great career advancements. Then OpenAI takes it, changes the colour scheme, finishes the login flow and claims the whole thing as their own.
Not only would it piss you off but it would also misrepresent what OpenAIs models are capable of.
Even in code, in person and online I’m seeing some reversal. It’s here to stay I’m sure but I also think “no one will ever hand write code again” is a narrative that is getting pushback.
It hasn’t been passed and no one cares about it because it’s basically an end goal. No lab can hit it so they can’t juice the crazy Turing benchmark 3000 for marketing.
If someone sat me down today with an LLM and a human and both were trying to prove to me they were human, and I can have conversations of arbitrary length, I’d get it right every time.
From the abstract: "When prompted to adopt a humanlike persona, GPT-4.5 was judged to be the human 73% of the time: significantly more often than interrogators selected the real human participant. LLaMa-3.1, with the same prompt, was judged to be the human 56% of the time"
Low n, time bound, not reproduced. And look at their example conversations…
And people forget that sometimes humans message twice. An LLM can only respond. So it immediately fails here in a true Turing test. (You could loop the LLM but then I expect even more immediately obvious bot behaviour).
A paper can’t reproduce its self. And he didn’t formulate the test with a 5 minute bound he just predicted that by the year 2000 that within 5 minutes an average interrogator would have a sub 70% chance of guessing correctly.
And it’s all irrelevant. If one human on earth can consistently get it right then it hasn’t been passed since clearly that human can somehow determine between them (whereas no one would ever be able to determine between a true “human intelligence” by definition).
And as it stands almost everyone could tell between them when allowed to discuss whatever they want for any length of time.
The fact these researchers have to keep adding bounds shows it hasn’t been passed. If we are arguing over technicalities maybe it isn’t as obviously intelligent as claimed!
> And he didn’t formulate the test with a 5 minute bound he just predicted that by the year 2000 that within 5 minutes an average interrogator would have a sub 70% chance of guessing correctly.
Right, the test duration was left unspecified. This means any duration is acceptable. Including, for example, the only duration actually mentioned by Turing himself in his paper. Or do you have a more authoritative source on which durations are acceptable?
> If one human on earth can consistently get it right then it hasn’t been passed
Says who? Not Turing. Probably he didn't say that because it would make the test both impractical and overly conservative.
> The fact these researchers have to keep adding bounds
The test was not "after thousands of hours of conversing with them, knowing they're AI, THEN see if you can tell them apart blindly." Were 2010 you to be in a real turing test with an arbitrary erudite human and a 2026 frontier LLM, not knowing LLMs existed, you'd probably struggle
This is always the most silly argument. The original test was ambiguous but for sure the human was trying to prove themselves human.
So the first thing they’d do is tell me LLMs exist and the other thing is an LLM. Obviously a true human level ai could explain that away as a fabrication to trick me. I don’t think an LLM could do even this!
Turings whole point was that through the medium of text along if the human and machine were indistinguishable then that was true intelligence. So yes conversations of arbitrary length are allowed (needed).
I doubt this entirely. It might be quite difficult for said human to discern whether a simple passage were generated sans such accumulated experience in reading AI text, true. But LLMs do not converse like humans in ways that have always been essentially immediately obvious.
Sometimes I wonder if LLMs are just revealing a section of the population with untreated mental illness or if LLMs are actively exacerbating mental illness.
We might eventually regret exposing the general population to such a new technology without almost any safeguards.
Have you seen how many people form one-sided bonds with stars that don’t know they exist? Lonely people suspend disbelief to find some comfort. It’s not proof that the chatbot is indistinguishable from a human companion.
So? The people who can do it prove it can be done - the people who can't don't prove the opposite. Might as well claim all math is wrong because most people don't understand it.
I like Musk companies more than most here it seems, I think he’s a net positive to the world. But (admittedly sans hard evidence) he absolutely gives me the vibe he’d train on my source code without my permission. I simply don’t trust him in this arena.
So there is more to lose than just the service being unreliable.
As a person he's either a poser or a neo-nazi.
As a government agent he's responsible for many deaths and as of yet untold harm.
Tesla was one of the earliest EV manufacturers, but these days they're overpriced and will be remembered for panel gaps and being destroyed by a car wash.
SpaceX has done a lot for that sector, but will mostly be remembered as the company that fucked astrophotography with 10s of thousands of satellites.
And Twitter has been dragged into irrelevance so badly it was being called a "reverse startup."
Neuralink seems to be going nowhere...
I can forgive him ransacking the federal government, exploiting workers in everything he does, building data centers that pollute like a cartoon villain, backing a fascist takeover of the US and similar ones all over the world, and openly sig-heiling, but I can't forgive him training on my source code!!!
I was using tabnine in 2020, I used copilot autocomplete + ChatGPT, then Claude Code. All this to say, I willingly adopted each wave.
Having said that, for all our startups “interesting” code I use no AI, not even tab complete. For a few reasons:
- having derived the code and built a mental model is something like 90% of the work, the code artefact being 10%. (You can probably verify this yourself, if you’ve implemented something once and then someone deleted it all, you could rewrite it MUCH faster the second time).
- once I have this model / vision in my mind, I’m streaming it from my mind into reality via code and having code suggestions pop up breaks that flow state for me.
- using Claude code even on throwaway scripts can mask problems. Why is doing this thing so difficult that it needs an AI to write the code? And usually when I’m writing those scripts I have a thought or insight about the system. in other words it’s nice to have that low stakes time.
- models are still frustratingly bad at spatial reasoning and I don’t see this improving anytime soon, which is a problem for a lot of graphics code.
- the code models produce still isn’t what I’d consider great code; subjective of course but it’s our product and we think having great code in our core technology is well worth it in the long term!
I do still use Claude code on various commoditised pieces (account UI, etc) and I use Claude as a great search engine, though its explanations of complex topics still kind of sucks and a human blog is almost always better.
If you are suggesting there is some new model of physics or groundbreaking technology that would allow such a dragless plane, then don’t let me discourage you! But AI won’t help at all, since it will only regurgitate conventional wisdom…
reply