FourWeekMBA

The OpenAI Drama

Section 230, Google Business Model, And The Evolution of The Generative AI Industry: https://thebusinessengineer.org/posts/the-end-of-big-tech

How To Redefine Your Career In The AI Era: https://thebusinessengineer.org/posts/moving-through-complexity

For a full picture, check this out: https://thebusinessengineer.org/posts/the-innovation-paradox

Read the full story here: https://thebusinessengineer.org/profile

Full description here: https://thebusinessengineer.org/posts/dismantling-google

Full explanation here: https://thebusinessengineer.org/posts/ai-moats-1

Listen to the full story of Silicon Valley with Federico Faggin: https://open.spotify.com/episode/2WkyQZmbbBzSUu7KSbXFNX?si=dsel-7bKRIeLocnHNBwb7g In this episode, we cover the following: - Neural networks, past vs. present - How human and artificial intelligence are fundamentally different - What's consciousness, and how it goes beyond classical physics - The limitations of AI - Is AGI coming? - How humans should make sense of this new AI revolution

AI Winter?

Is Google Search Dying?

Google vs. Microsoft: Google Advertising Machine, The New Google Search, Bard, BingAI, and ChatGPT

Is AI Getting Centralized?

Generative AI: What's Coming Next?

Read: https://fourweekmba.com/how-does-openai-make-money/

How Does ChatGPT Make Money? https://fourweekmba.com/how-does-chatgpt-make-money/

What other news is worth mentioning? - First of all, an incredible stat, ChatGPT might have reached 100 million users by January! For a bit of context, one of the latest successful consumer app, TikTok made it in nine months. So you can grasp the massive scale of ChatGPT adoption. - In addition, ChatGPT finally (and officially) announced the paid version at $20/mo (ChatGPT Plus). This is an interesting price point, as it shows that OpenAI wants to keep the tool, yes for B2B, but also enable it to become, potentially a premium consumer tool. Indeed, the pricing is not that far from a Netflix's subscription plan! Will it pull it off? - Another key point about ChatGPT's premium is that right now this Plus version is priced at $20/mo, but we might assume that OpenAI might be releasing a more powerful premium version with a higher pricing point, to tackle B2B. That segment, if priced well can become an incredible cash cow for OpenAI! - This week OpenAI also released an AI Detection tool. And I've seen a lot of people commenting how the game was over for AI content creation. That doesn't make sense to me, as AI content generation is a mouse and cat game. Of course, if OpenAI's ChatGPT is the only AI content generation tool out there, no doubt OpenAI has advantage if catching AI generated content. But otherwise, if the AI generated content can come also from other language models this will become a real cat and mouse game. In addition to that, even if ChatGPT is the only content generation tool out there, smart AI developers can still build various AI engines on top of those to make the content generated by ChatGPT indistinguishable from that of humans! Indeed, I played with AI detection in late December, and we also launched a tool here, which was slightly updated in early January. Yet, again, what matters when it comes to AI detection is the classification model, and that isn't something static, it needs to be continuously updated, as large language models get better, and as other developers build content engines on top of those large language models. So for those who believe to the results of AI detection tool religiously, you might be up for a great disappointment! Unless you'll build a company investing millions a month (as large language models become more and more complex) in AI detection technology, this will always be a cat and mouse game! - In the meantime, Microsoft seems to be moving fast in integrating OpenAI's technology into Microsoft's products. - This of course has awakened the sleepy AI giant: Google! Indeed, it seems that Sergey Brin, co-founder of Google was reviewing the code for LaMDA, the company's large language model (GPT-3's competitor), which might be the underlying model for Google's ChatGPT-like tool! - Indeed, Google has a huge amount of pressure as its revenue slew down substantially in the last quarter of 2022, and the only segment that made it strong was Google Cloud (which though runs at negative margins as Google is trying to win cloud deals). - In fact, as I explained, in AI business models, AI Supercomputers (part of the Cloud Infrastructure at Microsoft and Google) have become a key component to the AI race! So, if you are Google, you want to make sure to quickly fill the market gap between ChatGPT (which over time might turn into a Google's killer) and get back on track to the AI race! As this will help, not only, to keep Google's dominant position, but also to strengthen Google's Cloud segment, which in the future, might be the most important segment for the company and the infrastructure the will power up the AI Industrial Revolution! ​As I explained in yesterday's newsletter, today, ChatGPT is trapped into a web app, which doesn't access the web (for now) and it can't be hooked to your device (for now). And yet, once it does, with prompt engineering and in-context learning it might be able to unleash a set of custom experiences that we've never seen before. That might unleash...

AI Business Trends

Making Sense of The Current AI Paradigm

Visit: https://www.chatbusiness.ai/ How do you improve a model like GPT-3 if you are a small business owner? For one thing, it's extremely complex/or impossible to do it at scale. Instead, my main argument here is that, by simply verticalizing (creating a ChatGPT for every niche) this incredible tool, we can make it much more factual and grounded, of course, also more limited. In fact, one of the main issues of this current technology (large language models) is that they work (for the first time) extremely well on generalized-tasks. And if you were to restrain them too much, you would get. As a result, a much more constrained ChatGPT. So, as a way to learn, experiment, and limit the drawback of a technology/product like ChatGPT, we created ChatBusiness.ai, an AI business assistant, which has been fine-tuned on a crafted and curated Knowledge Base from FourWeekMBA. This, fine-tuning process, helps verticalizing GPT-3, while trying to limit its ability to allucinate, or give inaccurate answers. How? Well, let me break down the proces: - First thing is to make sure the AI model can be fine-tuned, on a set of highly curated data/content (Knowledge Base). - Then, you need to make it sure that the prompt, which is "the natural language coding interface" - if you wish - keeps the model grounded to the context it was given, while still being able to give a wide range of answers. - A last piece of the puzze is the reinforcement learning process, where you enable users to provide a feedback into each answers, and based on that prioritize again the learning and fine-tuning of the model. At least, as entrepreneurs, building tools on top of OpenAI, this is what we have control over, unless we build a foundational model ourselves (which might now might be a multi-million dollars endeavour). Thus, for now, ChatBusiness.ai is a limited version of the above, and if it gains traction I'll be investing more resources to further fine-tune, and leverage on reinforcement learning to make this a much better vertical tool for business people. In addition, for now, to keep things simple, this has been trained on the FourWeekMBA's knowledge base. Yet imagine, how, in the future this might get enriched from other knowledge bases of publishers that might want to opt-in. The interesting part? It is a traffic generator, in short, the tool speeds up the process of giving you the short, answer, to very specific business questions, and it sends you back the the article where this answer can be found. Thus, that is a win-win-win (a win for the user, a win for the publisher, and a win for the tool provider). Feel free to play ChatBusiness.ai, and beyond the feedback which you might be able to live when you get an answer, feel free to ping me back and let me know what you think! Keep in mind this is a first iteration, the interface is still quite simple and we're planning to further fine-tune it. But for now, you can start playing with it!