Function-calling is a powerful feature in Large Language Models (LLMs), enabling code execution in response to user queries1Semantic Kernel simplifies this by allowing built-in or custom plugins2. This article explores how Semantic Kernel creates a function-calling JSON schema, crucial for determining the correct function to invoke.
What if LLM and sLLM are just the beginning, and what lies ahead will be a breakthrough on the scale of the steam engine? And it will drive the entire global economy up. Do you think the guys selling horses, in the early 20th century, thought that Ford's cars would be accepted or become widespread ^^?
Truth be told, I feel like we're hitting some diminishing returns on information automation. Most the low hanging fruit is probably done now.
With ageing demographics and increased education around the world, I think the next major productivity gains will be in robotics - replacing all the manual labour and care giving that needs doing in one way or another.
LLMs may be a stepping stone in that journey but hardware cycles don't move like software and hardware adoption is slow and painful. This is part of the reason why VC doesn't like hardware but loves software
I think it's very difficult to predict even the near future with so many changes happening at once.
Consider even the previous excitement around the metaverse, that seemed to fizzle due to lack of content. Well now presumably generative AI can (soon) create all the 3D content you like, which could open many new possibilities.
All the virtual possibilities in the world doesn't reduce our burden to care for a growing number of old people. Nor does it reduce our need for food or other natural resources.
These will continue to be the bottleneck especially as we move to a deglobalised world, ageing and shrinking workforces in the developed world with population growth happening in EMs and FMs.
But robotics certainly will require more intelligence and better software. I had a Roomba for a while and that thing was just to dumb to be actually usable.
