Now having been two weeks since the release of my Chess plugin on the ChatGPT plugin store, I've decided to share some preliminary stats & thoughts. Initially achieving a day high of 1,800 games, we unfortunately experienced a user drop-off while the plugin was temporarily unverified. The plugin's return saw fewer users, likely due to an influx of new plugins. Despite this, I'm enthusiastic to see an average of 10 moves per game, with many players managing to beat the AI. The operating costs are minimal, mostly carried by storing secrets on AWS Secrets Manager. There's room for improvement such...
Over time, I've realized that my blog's front page looked a tad dull due to the absence of header images for some of the older posts. Hence, I decided to put AI-powered image generation platforms, DALL-E 2 and Midjourney, to the test. While DALL-E 2 offered some promising results, Midjourney stole the show with aesthetically pleasing images, despite the lack of API support and strict rules against automation. Despite these limitations, I managed to generate and download quality images from Midjourney, revamping my blog's appearance and learning a lot in the process.
In this blog post, I detail how I've made significant improvements to my long-time blog by integrating advanced AI tools. Using OpenAI's ChatGPT, I've successfully automated the tagging of my articles and generating precise summaries for each, enhancing my blog's navigability and readability. I also discuss how I used AI to generate related content, making my blog more engaging and interconnected. To top it off, I hint at an upcoming feature: AI-created images to beautify my blog posts!
We've seen two major shifts in technology trends with websites and mobile apps - now there's a third one rearing its head. OpenAI's ChatGPT with plugins is on the cards and you better be ready for it. In the midst of fumbling for answers to whether we need these plugins or not, let me reassure you that it's not too complex. Far from requiring a squad of specialist developers, all you need to know is how to make an API to create a plugin for ChatGPT. Yes, there are potential pitfalls around security and data protection, but with the right...
In another thrilling adventure with ChatGPT, I endeavored to integrate it with a Raspberry Pi, effectively turning it into my own personal light switch. My custom ChatGPT plugin, sporting just two endpoints, could revolutionize how we handle APIs, as it managed to control five different lights with simplicity and style. With a dash of YAML magic, ChatGPT managed to deduce the entire process, paving the way for future smart home projects. Big or small, the potential applications are endless and might even lead to computers communicating in human language!
Exploring the capabilities of ChatGPT, particularly GPT-4, I exposed a shortcoming regarding the model's ability to remember or store information it has 'thought' about during a dialogue sequence. Probing deeper, I developed an experimental system named ChatGPT Memory to input detailed information into the system like 'dreams', 'goals', 'inner dialogue' and more. While this method doesn't make the AI truly sentient, it definitely pushes the envelope and leads to interesting outputs. Although there are limitations, especially when handling more complex tasks, the enhancements present an exciting prospect for future iterations of the model.
Pondered the future of coding with a colleague of mine recently, as one does. Couldn't help feeling a tad underwhelmed by the programming tools like GitHub's Copilot X and ChatGPT. Sure, these AI-powered coding buddies are smart, helpful and do save time, but they more or less tinker around the edges - optimize the existing processes but don't fundamentally alter them. Software development still remains a tedious enterprise - a loop of problem-definition, code-generation, code-validation, and feedback. Not so different from what it was 30 years ago really. We're surrounded by subpar software everywhere - the bugs, the slowness, the...
In this enlightening blog post, I dive into the tantalizing world of ChatGPT and Large Language Models. Clarifying its operation, I unlock this enigma by comparing its mechanisms to a simple language model. However, Challenges arise due to the explosion of possible token combinations, leading to an inherent 'lossy' compression of our world's vast information. Surprisingly, even with such compression, these models can mimic human language in a compelling manner. I also investigate possible strategies to optimize this amazing technology - including zero-shot learning, one-shot learning, few-shot learning, and fine-tuning. Entering the era of prompt engineering and larger models, we're...
Strap yourselves in, folks, we're in for a wild ride! ChatGPT's new API has reignited my excitement for Large Language Models, just like the start of the dot com boom. With the pricing now 10 times cheaper, a flurry of creative and previously unthinkable use cases is within our grasp. Despite earlier doubts, I now believe we're scaling the peak of inflated expectations. Can't wait to see the innovative applications that will spring from this!
In a sea of ChatGPT chatter, we're stirring things up a bit with a cocktail chatbot of our own. From 'prompt engineering' to tailoring your bot for chit-chat about martinis, we've got you covered - the OpenAI playground is our bartender school! It won't just mix up your favorite cocktails, it remembers your preferences too, just like your favourite barman would. Be warned though: my fancy mixologist bot won't be held responsible for any questionable cocktails or spectacular headaches you end up with! So, pull up a barstool and let's dive right in.
Here's a wild ride with ChatGPT and other large language models! They're ace at cooking up plausible-sounding text, but they're not always the best when it comes to spitting out the truth - they've got a funky relationship with facts. One research paper showed they can come out with believable but totally fake answers to seemingly straightforward facts. But when I messed around with various models, there were a few discrepancies. Some got it right or plausibly wrong, but we humans are pretty gullible and tend to believe plausible-sounding info. So when it comes to using ChatGPT, make sure you...
In this article, I explore the different reactions that folks are having towards this tidal wave of AI in our lives, ranging from 'meh' dismissals to outright existential anxiety. I argue that AI is here to augment our capabilities, not replace us. While we might find AI occasionally stumbling, let's not overlook the amazing gift of efficiency and productivity it brings. Sure, we might be worried about the implications it has on our jobs, but isn't it also thrilling to have an AI companion working along with us 'cause if a computer can do my job better, maybe I should...
In the quest to explore the capabilities of ChatGPT, I decided to utilized it as a startup brainstorming partner. From product description to building the product on AWS and GCP, crafting an elevator pitch, highlighting the ideal customer profile, sketching a business plan, and even generating a logo idea, ChatGPT has been surprisingly helpful and creative. We even explored potential team structures that can bring the business to life! Turns out, ChatGPT might just be the co-founder you never thought you needed.