Neo Vision’s Weekly Rant #20
Where we RANT about what goes exceptionally good and wildly bad when you turn your company digital
The human brain works in mysterious ways. So mysterious that even now, 12.000 years after the Agricultural Revolution, we still don’t know much about it. But how did I get here? We all have memories that vary from blurry to really vivid ones. We even have false memories. One real memory (you have to trust me on this one) that I have, although not very detailed, is from when I was around 5-7 years old when I unplugged the landline phone and stuck the cable in my dad’s laptop to explore the not so Wide World Web back then. Most likely, I had no idea what I was trying to access, but the urge to immerse in that World was genuine. Fast forward 20 years, and here I am, absolutely stunned that I’m chatting with an Artificial Intelligence Bot educating me in seconds about any topic I desire.
So many things have happened in the last ten days since the release of ChatGPT that this newsletter will barely scratch the surface. ChatGPT is a chatbot based on the GPT-3.5 series of dialogue language models. ChatGPT was trained with the same method (Reinforcement Learning from Human Feedback) as its sibling, Instruct GPT, but fine-tuned using “Proximal Policy Optimization”. To understand better what ChatGPT is, some of its use cases, how it was trained, and what its limitations are, I recommend checking OpenAI’s blog post: ChatGPT: Optimizing Language Models for Dialogue. I must warn you that once you get down this rabbit hole, you won’t return. I will dedicate this week’s newsletter to ChatGPT by putting together people around the globe's experiences with this fantastic tool that got to one million users in just ten days. I will also introduce snippets of my conversations with ChatGPT.
Q: Who are you?
A: As I mentioned earlier, I am an AI assistant trained by OpenAI. I am not a person, but a computer program designed to assist users in answering questions and providing information. I do not have a personal identity or consciousness in the same way that a human does. I am simply a tool that is intended to be helpful and informative.
As always, Alberto Romero did impressive work distilling so many questions about ChatGPT, answering the most burning question: Are language models going to make Google obsolete? Of course, I asked ChatGPT the same question.
Q: Do you think that language models will make Google obsolete?
A: I don't think that language models, like myself, will make Google obsolete. While language models can answer questions and provide information on a wide range of topics, there is still a lot of valuable information that can only be found using a search engine like Google. Additionally, search engines serve a different purpose than language models, as they are designed to help people find information on the internet, rather than providing information directly. So while language models may be able to provide some of the same information that can be found using a search engine, I don't think they will replace search engines completely.
🤯 Frederic Besse, a Research Engineer at DeepMind, discovered that ChatGPT has unexpected capabilities. More precisely, you can build and run a Virtual Machine inside it.
Q: Why it took 70 years for people to be able to chat with computers?
A: The development of language processing technology has been an ongoing process for many years. There have been many significant advances in the field of natural language processing over the past 70 years, but there are still many challenges that need to be overcome in order to achieve truly human-like language understanding in computers. Some of the key challenges include the ability to understand context and nuance, as well as the ability to learn from and adapt to new information. Additionally, the development of language processing technology is closely tied to advancements in other areas of computer science, such as artificial intelligence and machine learning. So while it has taken many years to reach the point where we can have meaningful conversations with computers, I believe that this technology will continue to improve and evolve in the future.
💭 What’s the difference between traditional computing and AI’s natural language processing (NLP)?
Q: What was the inflection point of Artificial Intelligence?
A: It is difficult to identify a single inflection point for the development of artificial intelligence. AI research has been ongoing for many years, and there have been many significant milestones and advancements along the way. Some people might point to the creation of the first artificial neural network in 1943 as an important inflection point, while others might point to the development of machine learning algorithms in the 1980s or the rise of deep learning in the 2010s. Ultimately, the development of AI is a complex and ongoing process, and it is difficult to identify a single turning point or inflection point.
🚫 While many engineers are stretching ChatGPT’s capabilities to develop software solutions, Stack Overflow decided to ban answers generated by this language model
I use chatGPT the way I would have used stacked overflow in the past. My only gripe is how confident chatGPT is when it's clearly wrong sometimes - Kay, YouTube user
Q: What will AI programming look like in five years?
A: It is difficult to predict exactly what AI programming will look like in five years, as the field of AI is constantly evolving and advancing. However, some of the key trends and developments that are likely to continue in the coming years include the increased use of deep learning and neural networks, the development of more sophisticated natural language processing capabilities, and the integration of AI technology into a wider range of applications and devices. Additionally, there is likely to be a continued focus on improving the interpretability and explainability of AI models, as well as on developing ethical and responsible AI practices. Overall, I expect that AI programming will continue to evolve and improve in the coming years, and that it will become an even more integral part of the technology landscape.
✔️ I think there’s no better way to end this week’s newsletter than with what I call the ultimate ChatGPT uses-cases article: 🔓 Unlocking the power of the ChatGPT revolution: 100 💥 innovative use-cases to try before you 💔 are fired 🔥
Write me an ending for my weekly newsletter about ChatGPT
A: In conclusion, ChatGPT is a powerful and versatile language model that has the potential to revolutionize the way we interact with computers. With its ability to understand and generate human language, ChatGPT opens up exciting new possibilities for natural language processing in a wide range of applications. As we continue to develop and improve this technology, we can look forward to even more impressive capabilities and exciting new applications in the future. So stay tuned for more updates on ChatGPT and the exciting developments in the field of natural language processing!