Stop LLM research? Part -2/2
No diffusion models. Just PPT image merge with my own caption

Stop LLM research? Part -2/2

In the 1st part of the article I discussed the cost aspect and how quickly this technology has become prevalent at such low costs. This has caused good (mostly) and bad (it exists) actors to come up with disruptions which we could not have imagined.

In this post I will cover the 2nd aspect - ego blow to mankind. There were 3 blows to humanity's ego. To quote directly from Sigmund Feud

“Humanity has in the course of time had to endure from the hands of science two great outrages upon its naive self-love. The first was when it realized that our earth was not the center of the universe, but only a tiny speck in a world-system of a magnitude hardly conceivable; this is associated in our minds with the name of Copernicus, although Alexandrian doctrines taught something very similar. The second was when biological research robbed man of his peculiar privilege of having been specially created, and relegated him to a descent from the animal world, implying an ineradicable animal nature in him: this transvaluation has been accomplished in our own time upon the instigation of Charles Darwin, Wallace, and their predecessors, and not without the most violent opposition from their contemporaries. But man's craving for grandiosity is now suffering the third and most bitter blow from present-day psychological research which is endeavoring to prove to the ego of each one of us that he is not even master in his own house, but that he must remain content with the veriest scraps of information about what is going on unconsciously in his own mind. We psycho-analysts were neither the first nor the only ones to propose to mankind that they should look inward; but it appears to be our lot to advocate it most insistently and to support it by empirical evidence which touches every man closely.”

―?Sigmund Freud,?Introduction à la psychanalyse

Now, we have a bigger blow. AI is getting better than us in most tasks and it is improving performance exponentially.

No alt text provided for this image
Source: https://levelup.gitconnected.com/how-to-benchmark-language-models-by-openai-deepmind-google-microsoft-783d4307ec50

Now, it has become practical to deploy AI solutions to many human tasks, impacting a marid set of industries. Now most of us (including those doing AI research) are not willing to concede this point.

It will always start out with specific tasks for AI, but we tend to extrapolate its ability and then many tasks which need to be deterministic will be done by AI which is by nature stochastic.

To summarize - Copernican heliocentrism broke the myth that everything revolves around the earth. Earth was just another planet going around the sun. Then Darwin's theory of evolution proved that we (humans) evolved from monkeys. Then Freud said that human behavior is influenced by unconscious memories, thoughts, and urges - we are not complete master of our mind. Now LLMs are showing signs that is can out perform us on multiple benchmarks.

How are we going to react this sudden development? Obviously one line of argument is that - the AI is not really "intelligent" - its a clever Hans or stochastic parrot and we don't need to worry. But, its deployment and popularity hints at something more significant. We do need to worry.

#llm #llms #gpt #gpt4 #ai4good #aiforbusiness #capgemini #philosophy

要查看或添加评论,请登录

Rajeswaran V (PhD)的更多文章

  • Scaling laws

    Scaling laws

    A scaling law in deep learning typically takes the form of a power-law relationship, where one variable (e.g.

    1 条评论
  • Copy of GenAI/LLM and productivity

    Copy of GenAI/LLM and productivity

    I will present 3 papers which discuss this from economics point of view. The productivity J-Curve "THE PRODUCTIVITY…

  • Paper clip maximization

    Paper clip maximization

    There is an very interesting thought experiment called "Paper clip maximization" This is a thought experiment by…

  • AI and research

    AI and research

    Microsoft performed a lot of experiments with GPT-4 and released the results in the paper titled "The Impact of Large…

  • Moravec's paradox and CV

    Moravec's paradox and CV

    I want to discuss face recognition and how it fits in with Moravec's paradox. Background Steven Pinker writes "The main…

  • AI robustness

    AI robustness

    When we build AI systems - care should be taken to test its robustness. A decentralized group of safe streets activists…

  • AI for Software Engineering

    AI for Software Engineering

    For corporates, Software Engineering lifecycle is most important. This is most relevant for IT majors on where and how…

  • AI in 2024 - some predictions

    AI in 2024 - some predictions

    There is an old saying "Prediction is very difficult. Especially about the future !".

  • Dangers of over-simplification

    Dangers of over-simplification

    In 2021 Sam Altman wrote an essay "Moore's Law for Everything". It gives some insight into his thinking on how AI will…

  • LLMs and Theory of mind

    LLMs and Theory of mind

    In March when researchers in Stanford published the paper "Theory of Mind Might Have Spontaneously Emerged in Large…

社区洞察

其他会员也浏览了