The Deep North

[Done with programmer's assistants: Gemini, DALL-E] OpenAI's DALL-E produces images, but as webp files which can be awkward to work with. I have code here below for a web page that you can save locally that will allow you to select and convert a webp file. Meantime, here's the working conversion routine: Convert WebP to PNG Convert Download PNG OpenAI should fix this (as well as the response issues which has me using Gemini and Poe more often than not). Having to convert the image is a dumb workaround, but that is what you are left with. I have, in the past had to do various things with images, and although it can be a bit awkward if you are not familiar with it, ImageMagick is surprisingly competent with conversions and a vanilla type conversion is intuitively simple. Confusingly, the ImageMagick executable is 'convert'. To convert a webp to a png:  magick convert WhyThis.webp WhyThis.png ** With some installations, convert can be ca...

I am overdue writing up the fact that rapid AI advances have now made scammers much more dangerous. People don't seem to realize that things we thought were off in the future last year are here already and becoming so sophisticated so quickly they are, and will be, catching people off-guard. Until we have much better safeguards in place everywhere, you need to be very much on guard. Update passwords and put in place 2 factor authentication everywhere. A video call of a loved one can and eventually will be faked. If you don't know for sure that it's real, assume it's fake. This year you will have a very hard time knowing it's real. Unless you are expecting a contact from some source, you should start with a default condition that it is a scam. Scammers can now easily fake the voice of a person sufficiently accurate and undetectable that they can get into banking systems that use voice recognition as a security measure. An AI system can now tell if you have type 2 dia...

AI is Smarter Than You, But That’s Not Saying Much Artificial Intelligence (AI) has achieved remarkable feats, from passing bar exams and medical boards to acing SATs and scoring like a genius on IQ tests. However, it’s important to remember that AI doesn’t think the way we do. In fact, for some definitions of ‘think’, it doesn’t think at all. The Limitations of AI AI can make unexpected mistakes, such as adding extra limbs to animals in generated images or hallucinating facts and supporting references. These are errors that a human or even an animal with lived experience would never make. In the three-dimensional real world, objects cannot pass through each other like ghosts. While an AI can tell you this in text, it often violates this rule in practice because it doesn’t understand it holistically the way we do. The Risks of Relying on AI As AI continues to improve in areas like text construction, photorealistic image generation, full-motion video, invention, and other tasks we consi...

In my observations, well-intentioned individuals forming political platforms often fall into what I describe as "Grab Bag Politics," closely related to Pork Barrel Politics. This approach focuses on seemingly beneficial but ultimately minor issues, distracting from more critical matters. For instance, while it's admirable to want every child to have shoes they like, making it a legislative priority overshadows more significant issues, risking the neglect of broader principles like children's overall well-being and prosperity. Such specificity in legislation is impractical, as it's impossible to address every detail and maintain a holistic balance. Diverse priorities within a group aiming for societal change can lead to resource allocation conflicts, highlighting the need for a more generalized approach to ensure no essential aspect is overlooked and to manage priorities effectively. I advocate for a system that guarantees everyone a fair share of resources, maximi...

If you have yet to actually use an AI chat system, you are in for a surprise.  I will be writing up how the magic bit of GPTs relates to work I started more than thirty years ago. My work makes me think that current AI has found the magic I was looking for and that magic is very magic indeed. If what has been found is what it appears to be, there is much, much more magic to come, and it is almost certain to come much more quickly than people think. These systems are helpful already if you know how to use them, but they are going to be absolutely indispensable before the end of the year. It's moving that fast.  Here is the prompt I gave Poe:   Can you write quick step by step instructions for a new user to set up with Poe.com, and give a few examples an ordinary person new to AI would enter as prompts and what they could expect? Here is Poe's response: Certainly! Here are step-by-step instructions for a new user to set up an account with Poe.com: Visit the website: Op...

Quora paywalled my answer to a question.  As a high school junior with only basic coding experience, what should I do now to stand out on college resumes if my goal career is as a full-stack developer? You should do a project where you can learn and build something that demonstrates you are able to do something useful. You can show what you have done, and having muddled through it you can explain what you did. Put it up on GitHub so you can point somebody to it and to show that you at least know what revision control is. Note: AI is going to change everything fast so you need to add AI to your workflow. In this example here, use something like ChatGPT or Poe, etc. to ask/answer questions you need answered as you go along. When it comes to adding the database bit to your mini website, learn how to enter the code into an AI system and how to get it to aid you to write what you need. If you are organized and can walk through an AI session to go from this to that you can put up a blogg...

Below is a post that I made on Quora that Quora might make inaccessible due to their monetization thing, so...  Bob Trower I have been programming since 1976  Artificial intelligence (AI) is expected to become significantly more powerful in the future, with implications across various scientific disciplines and technological domains. Let's discuss the potential advancements in medicine, chemistry, physics, math, biology, materials science, as well as the progress in embodied AI, AR/VR, and cybernetics. We'll conclude by highlighting the power of AI in amplifying developer productivity and its own advancement. First, let’s have a look at underlying aspects. One aspect is the advancements of underlying hardware according to Moore’s law. Much has been said about how it must end, but it has not ended yet and with a movement to better technology due to AI (say, graphene processors), it is not likely to end soon. Also, I have overlain a line on the Kurzweil derived graph and you can...