"Helen has deep expertise in AI policy and global AI strategy research"
There is a funnel from academia to pluck individuals and propel them to policy level positions. None of the above's CVs remotely suggest deep expertise in anything but putting words together and being a "good soldier" for the Oxford-Rhodes thing.
-
"Americans should not be haunted by the specter of an imminent Chinese surge in LLM development. Chinese AI teams are fighting—and often failing—to keep up with the blistering speed of new research and products emerging elsewhere. When it comes to LLMs, China trails years, not months, behind its international competitors."
AI is not limited to LLM. The entire article is "nothing to worry about, China is behind in LLMs" which does not seem to serve America's interest, rather China's, and possibly NVIDIA and friends.
Even Foreign Affairs is going down the tubes. Somewhat depressing.
It never ceases to amaze me to see how often what is necessary becomes a necessary evil.
Dialogue is a necessity for the correct development of knowledge. To engage in dialogue we have recourse to language and a vocabulary of discourse. As knowledge is gained, dialogues branch off and specialize. The system of thought inherent to the ~unique arrangement of the elements of these dialogues partition thought into acceptable (an opening) and unthinkable (a closing).
Throughout history you will find minds that navigated these intellectual currents by stepping away from their dominant belief system and gaining knowledge of the universality of meaning, and saw hidden (filtered) vistas previously unseen, and then step back into their home grounds to contribute to the development of the field, in a positive manner, adding new elements into vocabulary of the dialogue. They extend it, and create the possibility of synthesis in the future.
Equally, you will find that the further necessity of establishing schools, cults, churches, and institutions, which lend social prestige to its members and satellites, introduce incentives contrary to that of pure love of knowledge, and this attracts a certain type of people, beyond the already present danger of vanity and self regard.
What is to be done about it?
Helpfully suggest that tolerance of this necessary evil may be the remedy for your gripe.
The picture captioned "Driving down Novinsky bulvar, Moscow" does anyone know the automobile brand from the hood ornament? Search engines spat out Ford Victoria or Fairlane from the '50s (text not image search) but the wings don't quite match.
> The brain is the single existing example of general intelligence.
This is incorrect. It is not pedantic to point out that we have never interacted with a "brain" in isolation: the human brain is an organ of the human organism. The human being is the single existing example of general intelligence.
> let's look forward to what neuroscience looks like in the 2030's
This is very interesting science without question. Are there existing ethical and moral frameworks guiding the development of your field?
If I was an intelligent person, I would use a basic phone for making phone calls and not carry surveillance devices on my person or have them in my house. This way I would not have to worry about keeping up with agencies that operate above the law, are accountable to none, and operate with huge budgets to subvert the ethics and mores of capable people.
> If I was an intelligent person, I would use a basic phone for making phone calls and not carry surveillance devices on my person or have them in my house. This way I would not have to worry about keeping up with agencies that operate above the law...
Why do you assume a "basic phone" would protect you in any way? It's far more likely to only be capable of insecure, easily interceptable forms of communication (e.g. SMS). Also, it's software is likely much worse than more popular phones (e.g. an egregious example is cheap Android phones shipping with malware preinstalled).
Have a really small 4g hotspot hotglued to a tiny Linux computer running the Tails distribution read-only with a removable SD card with all your data and no executable code on it if you're a real cypherpunk.
We really ought to push for something better than Tails. I'd love to run something like it on an aarch64-linux or riscv64-linux board. I'd love to run something that doesn't have a hacked, nearly broken debian boot process, which broke the ability to kexec it many versions ago, etc.
The 4g is in the hotspot that you're connecting to via wifi from the mini-computer. That way you don't have baseband firmware exploits to deal with on the linux machine like you would now with a traditional android phone. 4G firmware are all binary blobs that probably have backdoors.
So am I to understand that from an OpSec perspective, connecting a machine to a known compromised system, is ok to do, “because you want internet”?
Maybe because I’m not opsec and don’t know my ass from a hole in the ground, but my security recommendation would be, no, do not purposely connect your machine to a known compromised system regardless of its advertised purpose, attack vectors, attack surface, probability of unwanted exploitation, or justification as to why it’s necessary to do so, because you’re exposing yourself, and possibly corporate machine and network, to compromise. Find a trusted system (aka audited and considered reasonably low risk while acknowledging no system can ever be deemed fully secure and trust, or zero trust is a large determining factor) and consider the compromised machine as not existing at all, therefore not being an option at all, because connecting to it would go against common sense and 8th graders practice better security habits
I'm not sure what you understand to be a "basic phone", but they are easily intercepted and traced (triangulation from mobile phone towers, it's how the emergency services can locate you if you dial 112 all across Europe).
Communication has always been known to be a risky endevour with potential for various compromises, even for sovereigns. That telephony and now network communication infrastructure the world over is minimally at risk of, if not subject to, surveillance is understood.
Carrying a general purpose computer of substantial complexity that is equipped with state of the art sensors, optics, and components (including ai), which is then topped with yet another thick complex layer of software, which is subject to known and unknown access (by various parties), is not the same thing as a telephone (analog, digital, wired, or not).
Today, unless in a secured EM cage, there really aren't that many places where you can be certain you can have a private conversation, face to face. Visiting friends? Alexa and friends may be listening. Even the lousy TV sets :) Walk in the park? Your companion likely has a smartphone.
A healthy society requires the availability of private spaces and private interactions. When a citizenry becomes aware of pervasive surveillance it self censors. Self censorship prevents airing of views in an unemcumberred manner. When views are constrained, problems remain unaddressed.
Tyranny typically thrives in such insecure and non-optimal circumstances.
>Today, unless in a secured EM cage, there really aren't that many places where you can be certain you can have a private conversation, face to face.
Even then the assurance is only so high. Governments operate what are called SCIFs, Secure Compartmentalized Information Facilities, where they not only conduct physical exclusion and EM hardening but also acoustic damping so that an adversary can't, in theory, listen through the walls with a fancy stethoscope or a laser microphone.
There's a scene from Neuromancer where Molly and Case pay another character for a private discussion room, which is basically a cyberpunk SCIF. I've always found that scene oddly prescient; privacy today is quickly becoming a luxury.
> "- Battery would last several days on a single charge. I had totally forgotten that that used to be the case."
Several days? Surely you meant weeks? Unless you spent an hour or two calling on the phone every day, of course.
I get 5-10 days out of my iPhone depending on desire for leisure. In the "dumb" era the number was about 2-3 weeks, even when I was young and always texting throughout the day.
> I get 5-10 days out of my iPhone depending on desire for leisure
I probably use the phone a lot more than you. My iPhone has to be charged at least twice a day, and it’s not even that old. My current iPhone is an iPhone 14 Pro.
Yeah that definitely sounds like you're constantly on the phone - or at the very least keep everything and then some running in the background with app refresh and location services etc. enabled.
Any modern smartphone will have several days of battery life, if not more, if you use it as a dumb phone (WiFi & mobile data turned off, only using it for calls & SMS). My Galaxy S5 (~8 years old now?) passively discharged at 1% per day with WiFi & data disabled.
Not to mention that map application availability was equal to zero unless using the very latest breed of pre-smart phones from the late 2000s, and even then it was so-so.
> surveillance devices on my person or have them in my house.
Unfortunately, even if you trust your "basic" phone to not be compromised, it still means you can't have a personal computer. From what I've read, that's the strategy used on the Kremlin - for security reasons, they banned computers and went back to typewriters.
Are you implying that calls made from smart phones are anything but on the clear? Are you making all your voice calls using some merging other than the phone app?
Many calls are made using apps these days. WhatsApp alone had more than 15 billion voice minutes on average per day back in 2020.
If your phone and the phone of the other party are not compromised, it is indeed possible to conduct end-to-end encrypted calls with perfect forward secrecy.
> If your phone and the phone of the other party are not compromised, it is indeed possible to conduct end-to-end encrypted calls with perfect forward secrecy.
We have not done benchmarks against deeplake yet. Deeplake has some interesting concepts in their design, I'd be very interested to do a benchmark soon.
https://cset.georgetown.edu/staff/helen-toner/
https://polisci.columbia.edu/content/jenny-xiao
https://jeffreyjding.github.io/
Oxford and jaw-boning heavy, not an American in sight. How does someone like Helen end up being a director at OpenAI?
https://openai.com/blog/helen-toner-joins
"Helen has deep expertise in AI policy and global AI strategy research"
There is a funnel from academia to pluck individuals and propel them to policy level positions. None of the above's CVs remotely suggest deep expertise in anything but putting words together and being a "good soldier" for the Oxford-Rhodes thing.
-
"Americans should not be haunted by the specter of an imminent Chinese surge in LLM development. Chinese AI teams are fighting—and often failing—to keep up with the blistering speed of new research and products emerging elsewhere. When it comes to LLMs, China trails years, not months, behind its international competitors."
AI is not limited to LLM. The entire article is "nothing to worry about, China is behind in LLMs" which does not seem to serve America's interest, rather China's, and possibly NVIDIA and friends.
Even Foreign Affairs is going down the tubes. Somewhat depressing.