This article will come as no surprise to anyone who follows my articles or Threads.
I am a big believer that technology is rapidly moving deeper into the generative AI world. Whether this leads us to AGI or not is outside the course of this conversation.
What we are going to focus upon is the direction of LEO. For this reason, LeoAI is one of the biggest additions to the platform.
I also think it should be the next major priority.
Chatbot Technology In Applications
To me, it is a safe conclusion to state that chatbot technology, derived from LLMs, are going to be incorporated into most applications. Elon Musk is already doing this with X, as evidenced by the recent announcement that it will be available to Twitter Blue users this week.
Here is the model that Leo needs to follow.
Obviously, there is the desire to wait. We know technology gets better over time. There is also the leeriness of disappointing people since expectations can certainly go unmet. This is the nature of early innovation.
That is, however, offset by the need to get training. Rolling out the feature is a major boost to the value of the platform. It is also a starting point. Here is where we see the evolution beginning.
While few can expect it to match what we see from some of the major players, it will be a major step forward. Leo should take the lead in these areas. Bringing it to market, no matter how basic, is crucial.
The key is to get people using it. This creates the feedback loop that is common with associated technology.
Potential Of LeoAI
What is the potential of LeoAI?
This is a loaded question. Part of the answer is dependent upon where these technologies, in general, go. However, with LeoAI, we can start to lay the foundation of something that evolves into much bigger.
My view is that we are eventually going to see personalized AI assistants emerging out of LLMs. It might take a while to get there and it might be somewhat limited on a platform like Leo. That said, it is a direction we can work towards.
Leaving that aside, we can focus upon some of the basics of content creation. For example, we are discussing, so far, text. There is no discussion, as far as I know, of image generation. How about we have LeoAI generate an image when are creating an article? Is this something that could be integrated with the open source models that are out there? I would believe it is possible from a technical angle.
Could this end up moving into something like creating jingles or short songs that could be used in videos? Again, there is software emerging that can do this.
What about moving to text-to-video? This is really in its infancy and might be some time before any type of open model is available. However, could this be something that is available?
We already know the information provided by LeoAI, based upon the activity on-chain, is going to be unmatched. This alone is a handy feature, one that provides massive amount of feedback to the users.
One final thought: could we saw different prompts through our accounts? Perhaps there are queries that we do regularly which we want to store and reuse. Here is where a feature similar to bookmarks could be of benefit.
This has tremendous potential as a value add to the network. In my mind, it is something we want to get going as soon as we can.
A lot will be showing up across different social media platforms over the next few months. Leo needs to get this going.
Posted Using InLeo Alpha
That is a great idea and looking forward to it's integration in LeoAI.
LeoAI has a lot of potential, especially since it can access all the data in the blockchain. Those in Web2 can be hindered by paywalls, or other things. Reddit is even selling their data yearly. I guess the biggest hurdle I can see for LeoAI is the amount of data it is being fed. If it is just limited to the Hive blockchain, that is significantly less than its competitors. Data is vital in LLMs, and they need more of it. I also wonder how powerful the computers/servers are for their setup.