I’m leaving OpenAI to build telepathy
naomibashkansky.com
naomibashkansky.com
Yeesh. I guess that makes one of us. Truly one of the more horrifying short stories I’ve read.
Poe's Law at work.
>Thought-to-text, aka telepathy
No, that's not telepathy. I'd leave the term alone. Consider it to be reserved for... well... telepathy!
ever since Elon did his "promise full self-driving every year" thing, entrepreneurs been on a weird sort of fiction escalator.
thought-to-text is incredible. imagine doing something that remarkable with your life and feeling the need to exaggerate.
But sure, you're right. Reading is not a deterministic mapping of text to thoughts.
Apart from the trouble of defining what a thought is, independent from the person who thinks it.
Me thinking of grabbing a mug of coffee is a different thought from you doing the same... defining "thought" pulls in a whole jungle of epistemology (maybe I should say philosophy, more broadly, I'm not in the field).
Is there an abstract concept of "the thought of raising your left index finger"? Or "the thought of remembering severe pain"? Etc etc
When I write something, I deliberately curate my thoughts in order to convey an idea, often heavily depending on the context. Context encompasses the whole slice of the world that is connected to my individual life, who I want to address, and what I want to convey.
As a simpler counterexample: it's possible to write down lies, deceptively. The definition of a lie means that I know better than what I say.
So, "honest thought" != "fact":
where does this take us?
People who lie not only try to deceive others about facts, the core of most lies is deceiving about intent.
I used the word "idea" here, and I think it is nothing more than a rephrasing of Platonic ideas, but I ought to be careful, because we are more than two thousand years in the future from there.
It's been explained to me by neuroscientists I respect that EEGs and friends are like trying to listen in on a conversation happening on center ice from the parking lot of the arena. You can maybe tell if there is a match going on if there is a rowdy crowd but that's about it.
Technically fMRI is noninvasive; it is just a bit inconvenient for everyday use.
But, obviously those only make up a small number of people.
Strapping a mind-reading machine onto some detainee is probably the more likely case.
The problem is that anything based on LLM architecture is not going to be fast or accurate enough to allow someone to, say, fly a fighter jet with this interface.
1. Superhuman AI is inevitable.
2. Writing will become a bottleneck to communicate effectively with it.
3. Higher communication bandwidth would let us keep up with the machines.
4. Therefore with higher bandwidth humans could stay part of AI decision processes.
5. That makes it more “human”, as humans would be kept in-the-loop without becoming a bottleneck to be worked around.
6. If you let superintelligence write into your brain, that lets you merge with the machine.
> I put a band around my head, and open my laptop at Conduit. It’s 9 am. My device pairs with my laptop over Bluetooth. I open, without loss of generality, Codex.
At least we are reassured that short-range networking protocols will be invariant under the apocalypse.
Sounds magical
Ah, finally! To be liberated from the burden of speaking! Perhaps soon the AI will even think for us!
On a more serious note, this is additional evidence that Wall-E stands as the most accurate sci-fi projection of the future.
I cannot comprehend the output from AI fast enough for the speed difference between a verbal and thought response to matter.
Not to mention the privacy nightmare of logging my raw thoughts in a corporate environment and potentially routing them to an LLM provider.
Talking to a machine has the same effect on me.
Under current US law, can people entering the US be subjected to mind-reading?
So you'd need to force people to somehow think the same things they do when talking or typing, without actually performing those actions. If you can do that, why not just have them talk?
And, my point was, there may be zero words to collect at all, because they'll tend not to be actively thinking about any words.
https://en.wikipedia.org/wiki/Kyllo_v._United_States - The case I was referencing.
Again look at the run around on the fourth amendment or the commerce clause ruling that made it so growing wheat for your own consumption was legally considered interstate commerce.
The fourth is effectively gone in the digital age since so much activity is done via third parties who the government can compel to hand over your information.
And for the interstate commerce clause they twisted it to argue that completely _intrastate_ commerce was actually interstate commerce because you would have bought the goods from other states if you didn’t make it yourself[1].
I’m not even opposed to the ruling in terms of what the government should be doing, but you are being blind if you think that’s logically consistent. There’s no reason the constitution would specify “…to regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes.”
It would have just said they had the power to regulate commerce.
- still annoyed by AI-speak
What a bold vision!
When we have GPT-7 it will still do AI-speak?!
I read "race conditions" all over this vision.
The Conduit® Thinkerface® Pro Band will contain a small 90B dense local model that pre-screens your Conduit® NeuroTokens® stream for compliance against the Conduit® Acceptable Usage Policy applicable for your region and subscription tier.
If non-compliance is detected you'll be promptly notified via a small current applied to the Conduit® ThinkRight® Human Training Electrodes embedded within your band. Intensity and duration are dynamically adjusted until compliance is restored.
Even better, because this all happens on-device before your Conduit® NeuroTokens® are sent to the Conduit® NeuroServer®, non-compliant thoughts don't count towards your Conduit® Daily Thinking Allowance or expend credits from your Conduit® Additional Thinking Purchase Plan.
Unfortunately I think this is inevitable because as a species we perform a steady march toward collective desires for items of science fiction, so start your adversarial double-think practice now.
I’m generally pro-privacy, but here’s the hard truth: at my startup, we disabled keyboard input in our app Slopgen and switched the default to Conduit Thought Input™, and only 2% of users bothered to navigate the 17 submenus and submit their SSN and face scan to re-enable typing. Yep, 2%, that’s all.
DAUs only decreased by ≈11%, but our ARR is already up 170% thanks to our AdThink™ revenue sharing agreement, and only 1% of users argued with our Fin.ai bot for the 45 minutes needed give a thumbs down feedback about the change.
Let’s face it: talk is cheap but users *want* these features, revealed preference *always* tells the real story. Better get used to it now or get left behind.As for Utopias/Dystopias, I don't know how to answer that without waxing poetic about ideas like chaos being the natural order, or beauty being in the eye of the beholder, <insert other platitude>.
This is a Black Mirror episode coming to life.
The only thing I would actually want this for would be a dream logger. Extremely invasive and quite dangerous otherwise.
It doesn’t even really need to work of course.
We know that we can use classical techniques, (non-AI), to psychologically condition humans. In the lightest case, we do it with advertising and social media. I'm absolutely certain that state-level adversaries use even more effective techniques to train and condition agents for fieldwork.
If your adversary knows you are more likely to rely on intelligence gathered in this fashion due to its reliability, they can definitely game that. Or rather, I guess, I would definitely game it if I knew the Russians had this kind of technology. So I don't know why they wouldn't be thinking along the same lines.
Basically, make them think their new technology is working.. until it counts. If it consistently works 99 out of 99 times. They'll have no reason to believe it's not working the 100th time.
But I guess if you were interrogating, maybe a pot dealer or spouse muderer or cheating husband or something, it might be valuable.
“A couple years” Not “a few years” — “joyful extension” — that’s an optimistic prediction in two ways…
How come these brilliant machines can’t figure out how to make a hospital stay not cost $60,000? Or build a car for less than $30,000? Or keep a megalomaniac from fooling enough people to seize power in a country of 300 million people?
"I'm leaving OpenAI to build telepathy"
I feel new a trope is being born.
He starts off at 2027, where he’ll have a headband that can read his thoughts. That’s in 4-16 months. Correct me if I’m wrong, but last I checked we‘re quite far from being able to read people’s thought from a headband.
Do these clowns actually forget that part?
The statement implies that the company has some secret hardware that can, or at least, is close to doing that. That would blow my mind, but the author seems to believe it.
As an aside, I believe the author is a she.
No, they are aware. You and I are not the audience for this post. People with capital to invest are the intended audience.
> It’s hard to collect enough data using invasive methods. Few people want a chip in their brain! But non-invasive methods are getting much better. The hardware is improving and getting cheaper, though I apologize for being vague about the particularities of our hardware.
https://condu.it/thought/10k-hours
Their data acquisition routine was driven by Claude and they were recruiting random people off Craigslist pretty aggressively a few months back. I know a couple of folks who responded out of curiosity and described a very janky setup in the rented basements of a couple of residential houses in SF. From what I heard about the setup the data they got must have been incredibly noisy.
To the people who are sceptical here, I wouldn't count them out yet. They almost certainly can get funding with what they captured, which would get them better hardware. And given the state of brain scanning - people have captured images from scanned brain activity some years ago as I recall - I'd say they have a good shot. But yeah, it's not telepathy.
Previous HN discussion:
In a couple years?! Get over the technical hurdles, do you think PEOPLE will be ready to telepathically talk to a computer in A COUPLE OF YEARS?
How much are those people making to be able to afford things like this?
People like this are the input side of the "oops we built the Torment Nexus" pipeline.
What collapse of web 2.0?
Technically the website we're on is web 2.0 because we're interacting as users.
I don't think it's naiveté on their part to want to build new things. I think you're just jaded and hyperfocusing on the negatives.
Neuralink exists since 10 years, is in clinical trial, has been implanted on a living human being in 2 years ago.
The lack of adults in the room to set rules around moral hazard is hardly her generation's fault.
It is been 10 years ago when I looked through some research in that area. Back then it was the problem to get enough good signal. Has the situation changed radically here for non-invasive electrodes or portable detectors?
I understand that the approach would be to turn weak signal, context, and LLM based signal processing into useful human computer interaction. For everything else, I have the feeling we are still in speculative territory.
And this will provoke “well we need to regulate AI then”. No, you need to regulate poking around in people’s brains without asking them before that becomes normal.
"I am a former-OpenAI employee and looking for investors to fund my AI startup."
Can this technology also do text to thought?
Ah, yes, very joyful indeed.
Also, fuck those interns! They're just too smart to keep up with.
I'm trying to remember if I was as much a tech-optimist as she is, at her age. I remember fondly imagining a futuristic smart PDA (Personal Digital Assistant) that's everywhere with me and I can use it like in Sci-Fi movies and games.
I'm assuming this is roughly the same vision, for the next generation that already has smartphones?
Yet something about this new "modality" of listening in on your thoughts, hearing your brain, is just so icky and dystopian to me. Guess I'm old?
What a terrible idea.
Stealing all my brain waves is called non-invasive now.
> That means we must scale up our data collection by orders of magnitude beyond what has ever been done in academia.
What you can do you can always overdo. And if it hasn't worked yet, it's because we haven't overdone it enough.
However doomed-to-failure these brain-dead ideas may be, the sheer unscrupulousness is top SV. Mix in some thought control, and you'll have plenty of investors.
> Audacious yet practical vision: I believe in this vision; to quantify, if Conduit becomes the general read and write company, then I expect it to be worth over $1 trillion. I feel good about the vision, because it builds towards a more human future. And I trust the scaling laws, which have held for many doublings.
Well, good luck with that venture
Dead in the water.
It's like an omlette. Can't make a neural interface worth a damn without cracking open a few skulls.
I immediately thought, ironically, your neurology wrote that with a computer and HN humans' neurologies read it with computers.
If what's there in the bone stock I/O plane was sufficient, we wouldn't be having this conversation.
‘Tech’ culture has gotten very, very weird. It’s turned sour. It’s so far past the point where I no longer feel the joy and excitement and liberatory potential of computing. I’m not even that old. What happened?
It always had that edge. Maybe technology in general always has a somewhat pagan flavor to it (for example JPL/Jack Parsons). Its just becoming a bit more mainstream now due to AGI larping.
And yes, the company that owns this website has a lot to do with that. Just look at the borderline cult-like adoration for certain tools, or the uninformed comments by people who are clearly not SMEs in fields like medicine, aviation, or other critical fields about how AGI is right around the corner and will free us from the oh-so-sad fact that we have to interact with other human beings to live.
On the face of it, psychoanalyzing from afar, it makes perfect sense. Relatively asocial, unpopular smart kids who later grew to become kings of startup empires, who see human interaction as a net negative, who are quick to point out the fallibility of human intellect while simultaneously being incapable of acknowledging the fallibility of their computers which they have been glued to, who endlessly over-optimize and neuroticize and attempt to quantify human emotions and expressions which are famously unquantifiable and subjective, who see the intrinsic subjectivity of life as something uncomfortable and avoidable rather than something beautiful that we all must contend with, who were more than likely never regarded highly by their more conventional or perhaps less intelligent peers in primary or secondary school, who were never socialized properly, are now the kings of our new feudal realms. This industry promoted them to that position, and now we all must live with the consequences of that decision, for the better which decreases by the day, and the worse, which increases by every femtosecond.
Years ago I was close to disavowing technology and only doing the minimal effort needed to keep up with my job, because I'd gotten bored and disillusioned with it. Anything more than a small hobby project had accumulated too much complexity and busywork for me to feel excited about working on it. I suffered through it for the sake of the few things I cared about most, but it took its toll on me. I didn't have the infinite energy I may have had for my favorite projects in high school anymore.
But then this new wave of AI came along. First there was Copilot, and I was in awe not necessarily at its immediate practicality, but the potential it demonstrated. I felt the embers of passion tentatively peek their heads out from wherever I'd previously buried them.
Every new leap, computing became more and more rewarding again.
First general searches became incredibly easier to find the relevant bits I wanted from. I actually wanted to look things up again now that I didn't have to grok dozens of pages of Google or Stack Overflow. I got my custom home router configured the way I wanted it instead of giving up or settling for copying some known config that wasn't quite what I wanted, after a small amount of trial and error.
Then it became reasonably competent at simple but tedious tasks, at least 90% where the last 10% was easy to fix by hand. Enormous git conflict to slog through? Instead of staring at the screen feeling my heart sink and wandering off to bedrot instead, I felt excited at the prospect that I could instead command the computer to do it for me, and it was actually motivating to get that to work. Instead of feeling too exhausted and annoyed to keep working, I felt rewarded and refreshed.
Then it started writing entire code files, then entire features, then entire applications all on its own, with my guidance and input on architectural decisions and aesthetics. Performing entire FreeBSD upgrades while keeping custom kernel module packages working, fixing Gentoo systems with messed up USE flags and an otherwise intractable dependency hell to sift through, hell, even writing entire kernel patches to fix sleep mode and graphics problems I'd been having on my 2010 Macbook Pro with Gentoo installed. I feel something reminiscent of my memories of opening Christmas presents as a child when the AI is done making something I'm super excited for.
I'm eternally grateful for the current AI zeitgeist in the tech world. We're living in the best time so far in history to be into tech. And things like Conduit? How would you not be excited for something with so much potential to be made? Even if you don't like it, imagine what it could mean for people with communication disabilities, carpal tunnel, etc. I'd love to be able to just type with my thoughts, and sending them straight into an AI sounds awesome too.
If this is satanic then I don't want to go to heaven. I feel more joy, excitement, and liberatory potential in technology than I have at any point in my life. I even feel more liberated to learn more - without the tedium raising my cortisol and stealing my attention all day, I'm actually more inclined to do things like write code in vim on my VT320 or read over code there with no distractions or AI input. I've reaped great rewards from being able to think more clearly and at higher levels of abstraction, free of the bonds of bullshit and tedium, at peace in the knowledge that I will never have to shave a yak against my will ever again.
Talking and listening are a really still a big mystery. And thinking is deep underground beneath our understanding.
We barely understand what we are doing when we are doing any of this. Personally I think I under stand what I'm doing when I'm writing much better than what I'm doing when I'm listening.
I'd find that a little surprising, except I've been doing it all my life. I barely have any memories from before I talk, and I have loads more memories from after I learned to read.
Learning to draw and play a musical instrument deepened all this. As did having good, close friends, and partners. I expect having children, which I haven't done, is even more deepening.
The thing is, all these words are just that: words. Very loosey goosey and they change from person to person just as the communicators come from different experiences. And tomorrows me often has a very different outlook than yesterday's me.
I'm not surprised that I can't read the mind of tomorrows me but I'm always a little shocked when I realize I can't read the mind of yesterday's me.
What was I thinking!
I can easily read that blog post as a description of what might come to pass. It might very well not connect very well with us now because, well, we haven't had the experiences yet. Telepathy might mean something very different after we've had experiences with the new technology.
My grandmother, who was born in 1893, often told me about how the world changed radically and almost imperceptibly during the first half of the 20th century. She lived to be 97, and she once said she wondered if would be possible to talk about the, then, present to her younger self. The words she used at the time of the conversation, would not communicate very well because of the lack of common experiences.
I think Shannon had something to say about that. Information is something new and unexpected, but you need the expected to frame it. Give it a foundation.
I'm now in my late 60s, and I can totally relate to what my grandmother was saying. The main difference is that we are now accelerating much faster.
2026 me would have a hard time explaining things I'm doing with AI to 2023 me.
I think that 2027 might be a bit early for some of the things the blog post is describing. But, come the day, I might just shake my head and wonder what in the heck I was thinking.
Ask anyone over fifty how far away 2035 is and they will tell you something like it's not long after Christmas.
I am fully willing to stipulate that Naomi believes it earnestly (rather than being like Musk just kicking things down the road) but those are very, very optimistic scenarios.
> My AI is a natural extension of me. It feels like a sixth sense and another limb. I wonder about a problem, feel as though I’m literally surfing the web, see glimpses of the websites, get flashes of intuition about the problem, and ultimately derive the answer.
We call this AI psychosis.
The article's topic is brain to AI interfaces. She mentions this quote as a "vision" from 2035, where not only can thoughts inform AI but AI can respond in thought too.
When we get used to a good tool it becomes an extension of ourselves.
How is this AI psychosis? Is it that it's very unrealistic for such an interface to be created in the next 10 years? Or is it that there's a fear of letting AI interfere with one's thoughts lest they lose themselves?
I think it's just Hanlon's Razor getting dull. The psychosis is the lazy malice disguised as a clever one-liner, and as increasingly frequent around here lately, with no substance, of any kind. As evidence for the dullness, just look below and observe what happened to the other question asking the same thing.
I think it's time someone developed a piranha, fishbowl-themed wrapper for HN, where the visuals would better convey what's actually happening here. And it would fit the cartoon level depth many are seeking.
Perhaps this is also why they're so worshipful and quixotically obsessed with the vaporware, God they call AGI so much, because they cannot understand how competence at tasks is not equivalent to intellect and that intelligence as we understand it is not one-note nor is fundamentally inseparable from certain sensory experiences that computers cannot currently have. The fundamental underpinnings of AGI or LLMs prevent that from occurring in the near-future unless you literally set out to create RoboCop or a T800. We all know how that ends.
Most of your life is spent performing a business function; the value of your labor is rapidly declining, and the little freedom the common man has is evaporating. Further, the surveillance state is near total - everything we do is recorded, and these records are controlled by the worst amongst us(including this message from me to you).
This is not intended to be provocative. I am asking sincerely. I have a lot of AI sessions, that I value enough to have reviewed multiple times, over months. And indeed, I explored much that I never would have without the AI, and enjoyed it. Even appreciated it. The bulk of my saved material is research, but at least a gigabyte is wild speculation, philosophy, and assisted thinking.
At what point does what the user thinks is a meaningful experience become psychosis, or is it that exactly; attributing meaningfulness is the symptom?
Edit: The "no longer other" statement, I should admit, is familiar to me. There can be, on rare occasion, what I have thought of abstractly as harmonic resonance, some kind of strange emergent state between two interlocutors. Maybe that is the psychosis?
Just because an AI bot says things that are meaningful, that doesn’t mean you need to assume it is intentional, sentient, or intelligent. It could vomit out a random Rorschach test and if you found meaning for yourself, you can still understand it was random.
I talk to a bot to sharpen my thinking, not to adopt its thinking.
Right. I was under the impression it doesn't have any thinking. But that never mattered to me much. And I would expect someone working with OpenAI to have a firm handle on that.
I appreciate the effort to explain, but I kind of took all that for granted. I still don't understand where the psychosis comes from. If the model spits out a number, the number is not false simply because it had no agency to understand the number. If the model brilliantly patterns something, I do not need assume intent.
If by psychosis is implied the obvious, such as attributing consciousness to an LLM, I might understand, though I would argue it's pretty elitist and self-righteous, and that "naivety" would be a less damning and more appropriate term, which could be used just as effectively. I am beginning to suspect it's a term that appeals to a younger generation, where cruelty is a bit more normal than it was in my generation. Not sure though.
This applies to math discoveries as well. The Jacobian conjecture was always invalid, only our human understanding of it has changed thanks to human inquiry and validation, despite LLMs doing the hard work.
I feel like we are in that narrow window of time when humans grapple with a new technology by personifying it before we realize it is just another tool humans have created to amplify human capabilities.
But to me what you are describing is pretty close to it. Because we know there was never any harmonic resonance and or "strange emergent state between two interlocutors".
I say this as somebody who uses the frontier models everyday for work. They are impressive and useful, but transformer intelligence is not at all like human intelligence, even if it superficially resembles human thinking. That should become more obvious the more you use them, not less.
What parent is describing is more akin to believing in astrology and tea leaves.
Is the difference here between us that you are using a passive-aggressive drive-by snipe, probably to preserve ill-gotten karma tickets? I will suggest you may have been the cause of the illnesses you cite. You have an abusive quality, which is known to exacerbate such disorders. As a coward, you may have pushed things too far with these 'unfortunate' people, and found through a bit too much expectation -- if not encouragement -- what you quietly desired to find. Similar to what you have done here.
The truth is, those, such as your self, who recklessly pathologize others recreationally, often cause far more harm than the supposed crazies ever do, even to themselves. And for the gullible people here, as you flag this comment and become pleased with yourselves -- before doing so, ask yourself: Is "harmonic resonance" used in a rinky-dink, heavily moderated and filtered forum thread truly adequate grounds to start issuing psychological evaluations? I know you guys pretend to value ethics and fancy yourselves as sentinels of better rationale, but I think you'll have to live with something on your conscience if you don't pause to think about this one before the instant gratification of dragging someone down. Retrospect might point the finger of pathology in the other direction...
Pick your fights wisely. Have some integrity.
I'll also say the alliance you've formed with the little green man is touching. You've just met and it's already using "we" as if you're in a special club together. How adorable. "we know that it can lead there with AI." I'd love to see your club share your concerns with Yampolskiy, or some of the folks that actually work on frontier models, or any of the well-respected people who take the subject a bit more seriously than you suggest is healthy.
Is it really inappropriate to say, if sincerely and without exaggeration, that you make me nauseous? You do.
Without commenting on how realistic it is, I will point out that if you took the average person on the street taking a call on their bluetooth earbuds, and send them back just ~2 decades, they would look like delusional people talking to themselves.
I suspect a lot of the weirdness we see today would be normalized in the future, even if somewhat annoying (like BrainRot.)
The AirPods of today are pretty much invisible which is why it’s genuinely impossible to tell even with a close look who they’re talking to, especially when they stare at ME because I’m staring at them trying to figure out if they’re talking to me!
Please apply your great psychological prowess to Roman Yampolskiy or Tegmark. Those might require more than a cheap comment to leave uncontested by the spineless swarm here.
Go on. Make your case. I can still read it even if this comment is flagged again.
If you're in northern California anywhere near the Bay Area, work in tech, and are reading this, I am literally begging you to leave your echo chamber. This isn't healthy or grounded in reality in any way, shape, or form. It's not just the AI stuff, it's the whole mindset. Humans aren't meant to operate in that sort of environment where there's unlimited funding and no consequence.
When you reach that level, you have to work incredibly hard to screw things up.
One again we get the lamest version of a dystopian future and all you need is a black mirror in your pocket hooked up to a very smart parrot to become insane.