What ChatGPT really meant?
#ChatGPT has taken the world by storm, no doubt! But the lessons people talk about wrt what it means is mostly misinformed.
👉🏽 Open AI’s GPT-3 based models aren’t the best performant or most accurate in the world today, and it doesn’t do anything useful for any critical business function today. Yann LeCun, Chief AI Scientist at Meta explained his skepticism of how much of a breakthrough ChatGPT really is vs. simply an effectively marketed distribution of a well-known paradigm in AI (https://lnkd.in/g6ezdYxA)
👉🏽 And No, this is not a sign that #Singularity is near or #AGI is here. Far from it. That said, we are still in the early days!
But here is what it may really mean:
1. Innovation requires input from the edges — customers, users etc — ChatGPT proved right the fundamental startup tenet — Ship first, and ship fast! It’s the only way to learn about your customer and deliver value
2. Conversation is the ultimate User Interface — the simplicity of a conversational search bar (stolen from Google) with contextual capture, and maintaining the history of conversation was it needed to capture the imagination of millions of users. Vishal wrote a fabulous piece on this: https://lnkd.in/gHS64mG8
3. “No force on earth can stop an idea whose time has come” ― Victor Hugo — AI has been an idea and has been evolving for a while, but by getting into generative tasks and smart distribution of this to the masses, AI goes mainstream!
4. Foundation models like GPT-3 , DALL-E, BART that are generally capable of performing multiple tasks, will lower the barrier to entry into using AI/Machine Learning in your processes and tasks. These become the new abstraction layer for building intelligent applications for 90% of the world, RIP classical #MLOps
5. There will be a lot of noise, but that is expected. #AI is transformational to humanity and don’t lose faith, but a new breed of snake-oil sellers will emerge with #GenerativeAI and #ChatGPT and it will be hard to discern the noise from the signal. I’ll be doubling down on driving clarity and education via Stories in AI.
6. The risks of using these models without considering the risks and limitations is very high, especially in areas like #healthcare. Abhishek Gupta of Montreal AI Ethics Institute writes a deep dive on the risks and challenges with LLMs, especially #chatgpt — fascinating dive into the reality — take a moment to read it https://lnkd.in/eBRCHMB4
Like everyone else, I’m fascinated by what this means to a future. It’s no doubt that we we will live with intelligent systems and #AI powered processes.
ChatGPT and OpenAI is owed immense technology-powered gratitude for bringing #AI back into dinner table conversations.
What do you think? What does this mean to you and your business?
