This nugget from our Gen AI research study shows why it’s essential for
@Google to get search/Bard right. Using as more capable search/knowledge base is the singe most common use case of Gen AI today.
People’s primary expectations of Large Language Models: looking up facts. VS what LLMs are innately good at: creativity.
The brain is a prediction machine. We make bad predictions all the time. What keeps us from going into lala from a bad prediction is the ability to update those predictions in real time
Some interesting observations on AI LLM
Great quote: “An LLM is 100% dreaming and has the hallucination problem. A search engine is 0% dreaming and has the creativity problem.”