1. The healthcare and health insurance marketplace is riddled with regulation. There is no such thing as offering a health insurance plan or a healthcare service in the United States absent any regulation. Every existing service is, more or less, imposed via regulation.
2. The current health insurance system dates, in its current form, either to shortly after WWII or to the adoption of the ACA, depending on your perspective. That’s somewhere between 14 and 75 years. Where are the plans you speak of from these entrepreneurs? What are they waiting for, letting all of this sweet, sweet, arbitrage pass them by? Biding their time? Toward what end?
Disagree. This feels more like the Google partnership with Apple' Safari that has lasted for long time. Except in this case, I think is OpenAI who will get the big checks.
This integration is way more limited and frictioned. Whereas with search Apple's fully outsourced and queries go straight to your 3rd-party default, Siri escalates to GPT only for certain queries and with one-off permissions. They seem to be calculating that their cross-app context, custom silicon, and privacy branding give them a still-worthwhile shot at winning the Assistant War. I think this is reasonable, especially if open source AI continues to keep pace with the frontier.
If Apple wasn't selling privacy, I'd assume the other way around. Or if anything, OpenAI would give the service out for free. There's a reason why ChatGPT became free to the public, GPT-4o moreover. It's obvious that OpenAI needs whatever data it can get its hands on to train GPT-5.
ChatGPT was free to the public because it was a toy for a conference. They didn't expect it to be popular because it was basically already available in Playground for months.
I think 4o is free because GPT3.5 was so relatively bad it means people are constantly claiming LLMs can't do things that 4 does just fine.
One of the first ones I can think of is pairing it up with Browser extensions to increase productivity of knowledge workers. Think of sales people quickly assessing the viability of leads, or simply reducing noise on social media as a B2C app.
What I found interesting was the fifth section that describes the strategies to improve RAG performance. Basically:
1. Quality control. What documents to include?
2. Timing. When to query?
3. Pre & post processing. Improve LLM outputs based on retrieved data.
4. End to end training. Expensive, and data intensive but possibly the best long-term approach.
5. Controller. An interesting idea with similarities to Reinforcement Learning.
I wonder what are your thoughts. Which one is most promising? What has been your experience when building RAG apps? Also, is RAG the leading architecture for building applications?
