Thanks for the source so I can put this into context (which is the context of Russian disinformation, not grassroots beliefs representative of the anti-vax movement).
I think you are (subconsciously) strawmanning the anti-vax movements like Querdenker. Most of these believe that mandatory vaccination (or reducing freedom of unvaccinated, or making it economically infeasible/required to work) is bad and goes against individual human rights, and that the risks and benefits of vaccines were not clearly communicated.
So, even if you did not make it up, it is twisting the viewpoints to reduce their legitimacy by tying these to ridiculous theories. One could do similar by cherrypicking vaccine proponents and their ridiculous theories (like claiming COVID came from the wet market).
If these channels are not indexed, I have a hard time believing you, given your misgivings and ridicule on your other statements. If a discussion about "Pandemic was faked to get children out of caves" can be sourced, please do so.
AI output is already more careful and fair and balanced on these matters.
> "Everyone experiences hallucinations," Anil Seth, a neuroscientist at the University of Sussex in the UK, told DW.
> "It's important to recognize hallucinations can come and go during our lives at points of stress or tiredness," Seth said. "There is a bit of a stigma around hallucinations. It comes from people associating them with mental illness and being called crazy."
> But it's actually very common and happens even daily. The itching Yarwood experiences is particularly common, especially after drinking alcohol.
> "It's also common for people with reduced hearing or vision function to get hallucinations in that ear or eye," said Rick Adams, a psychiatrist at University College London. "These are non-clinical hallucinations because they are not associated with a psychiatric diagnosis."
Around 12k fatal outcomes have been reported in the EU after vaccination, but it is not certain in all cases that vaccines were the cause.
The vaccine tracking chips come from two Microsoft (-affiliate) patents, one about using chips to track body activity to reward in cryptocurrency, and another about putting a vaccine passport chip in the hands of African immigrants. That vaccines contain tracking chips is a fabricated conspiracy to ridicule and obfuscate.
Lizard people is often an anti-semitic dog whistle.
Rich elites use blood transfusions of young people to combat aging and age-related disease.
Children have been kept in cages and feral children have lived in caves.
You likely made up the part about faking a pandemic to get children out of caves, unless you can point to discussion about these beliefs.
Current AIs are RLHFd to avoid being a "yes man"/sycophant.
The point about employing better prompting is well taken. Don't ask "Who was the first female president?", ask "Was there ever a female president?". Much like on StackOverflow you want to ask the right question and not assume things (since you don't know enough to make assumptions).
Imagine if every time on early Google you found a spam result and then blame the search engine for that (and not your choice of keywords, or ignoring that you always want to return something, even if remotely related). Like a user banging a slab of concrete with a chisel and complaining that this does not produce a beautiful statue.
"confabulation" could be "solved" when LLMs realize they are uncertain on a reply and making things up. But that requires the LLM saying "I don't know" and rewarding that more than a wrong guess. That requires a change in loss functions and not even sure if all users desire that.
You've seen the demos of a couple holding hands and walking, or the museum shots where all the paintings maintain coherence, or a woman temporarily obscuring a street sign. Or you haven't seen those demos. Either way...
I feel LeCun got roped in debating the likes of Marcus and Yudkowsky. This has made his arguments lose nuance and become rigid. I also can't escape the feeling that if Facebook was tuned into Transformers, they would have shipped earlier, so there must have been some resistance or underestimation that's now repeated "They can't reason", "They can't plan", "They can't understand the world", "They are a distraction / side road to AGI".
It is kind of ironic that researchers who claim LLMs lack adaptive intelligence seemingly refuse to adapt their intelligence to LLMs. If even GPT-3 can find logical holes or oversimplification in your arguments about GPTs, at one point this starts becoming embarrassing and unbecoming.
> The generation of mostly realistic-looking videos from prompts does not indicate that a system understands the physical world.
While arguably true, it also does not indicate that a system does not understand the physical world (reflections, collision detection, gravity, object permanence, long-term scene coherence, etc.).
If LeCun wants to argue it does not understand the physical world, he should do so directly. Not attack something that is not directly stated, but rather convincingly and tentatively demo'd (I myself find it hard to argue that a system that generates novel pond reflections has not memorized/stored in weights some generalization program to apply to realistic scene generation).
This demo shows it is not even a wild prediction to guess that soon (consumer tech) we will be able to discuss visual scenes with conversational AIs.
> If we assume that every intelligent species will turn the universe into goo
The analysis made in the article is of similar bad form. If I am arguing a meaningless point, I suspect that is due to the all-round meaningless points I am responding to.
> if there is a hope that humanity will become grabby aliens (Star Trek).
You talk as if being a loud grabby alien to other space civilizations is a hopeful thing for humanity to aspire to. Did not Star Trek had a Prime Directive for some reason? Though I do like solving the Fermi Paradox by humanity taking a good hard look in the mirror. May disprove us being Big Foot Medusas with one stone too.
Humanity just got rid of a nasty period of colonizing space on earth. I would not let us near intergalactic space without a time-out period.
This theory is itself grabby, and grabs itself at that.
> Furthermore, we should believe that loud aliens exist, as that’s our most robust explanation for why humans have appeared so early in the history of the universe. But if loud aliens will soon fill the universe, and prevent new advanced life from appearing, that early deadline explains human earliness.
Even furthermore, we should believe that Big Foot Medusas exist, as that's our most robust explanation for why humans have not seen these yet and are still alive. Since Big Foot Medusas turn everything alive into stone upon observation, and we are still here, it suggests we are early in our explorations and observations of Big Foot Medusas.
Anyway, a loud grabby alien civilization expanding is "first come, first serve". Perhaps we have not seen grabby aliens, because the first grabby aliens became quiet, and make new grabby aliens impossible, or they finally became superrational and realized they need more than one civilizations to have an economy, cooperation, cosmodiversity, or competition: If aim is to win at tennis, you can't play tennis alone.
As for earth specifically, it was probably already grabbed, and humans are the terraforming organisms put in service of the grabbies to make our planet habitable. Or the grabbies are in such a zeal to expand and get the most of our galaxy that they focus on planets that are close to the horizon, after which they will be moving away at faster than light speeds due to accelerating expansion of the universe. Only after that will they get around to Earth (first building a Dyson Sphere around the universe to harness its energy).
If there are guardrails in place not to output sensitive data (good practice anyway), then how would this technique suddenly bypass that?
I still have trouble seeing a direct threat or attack scenario here. If it is privacy sensitive data they are after, a regex on their comparison index should suffice and yield much more, much faster.
I suspect ChatGPT is using a form of clean-room design to keep copyrighted material out of the training set of deployed models.
One model is trained on copyrighted works in a jurisdiction where this is allowed and outputs "transformative" summaries of book chapters. This serves as training data for the deployed model.
If ChatGPT ever outputs other user's chat history, the company is as good as dead. If that could be exploited using this technique that is out in the wild for over a year: show me the data.
Then again, if you have access to a model trained on sensitive data, why not ask the model directly, instead of probing it for training data? If sensitive data never is meant to be reasoned on and outputted, why did you train on sensitive data in the first place?
Good point. But many are already directly training on output from GPT. Probably more efficient than copying the raw training data. Especially if it relies on this non-targeted approach.
That is a pretty convoluted and expensive way to use ChatGPT as an internet search. I see the vulnerability, but I do not see the threat.
I've seen it "exploited" way back when ChatGPT was first introduced, and a similar trick worked for GPT-2 where random timestamps would replicate or approximate real posts from anon image boards, all with a similar topic.