Signal 6: Social-AI

Takeaway:

You can now train an AI chatbot with your own knowledge using the ChatGPT API. This means we will have many personalized AI-friends, or rather a chatbot in every app. The results of accelerating technology will most likely not be super-human, but extra-human. Outside of our experience, but not necessarily “above” it.

Data scientist, Izzy Miller downloaded 500,000 messages from his seven-year group chat, then trained an AI language model to replicate his friends wrote in his blog post: “There’s something so delightful about capturing the voice of your friends perfectly. This has genuinely provided more hours of deep enjoyment for me and my friends than I could have imagined.”

It shows just how far and much we’ve come in terms of accepting this new form of AI-human relationships. This is just the beginnings of social AI.?

So, what will happen next?

Training tools are becoming more accessible. Businesses can use the ChatGPT API to train an AI chatbot with customized knowledge base by fine-tuning the model on their specific dataset - each specialized for a specific task:

• A student could have an AI trained on their semester's readings, lecture transcripts, and notes.
• A chef could have an AI trained on their recipes, entree sales history, current ingredient inventory, and the menus of hundreds of rival restaurants.
• For news, one could imagine an AI trained on the complete archives of The New York Times.
• For innovation in agencies: we could have an AI trained on our marketing experiments, hypotheses, and results. AI can be trained on large datasets of?creative work, such as advertising/audio campaigns, design projects, or?music compositions. By analyzing this data, AI can generate new ideas and inspiration for creative strategists to explore.

What might Bridgeneers do?

It’s 2030 and “Doing your taxes” is going to be a nostalgic reminder of the olden days when AI agents didn’t simply do your taxes. Think what the future might look like and move in that direction.