Thanks for this excellent summary and critique of the Altman interview. I find Altman to be a bit of a hype man (understandably), but I do think the business opportunity that no one is really talking about is the effective 'retrofitting' of AI onto legal companies. Someone is poised to make generational fortunes here...
What I’m really waiting for now is AI’s ability to make logical, real-life decisions not just fast or intelligent outputs, but choices that actually make sense in context. Yes, memory is improving and systems are starting to retain more, but AI is still moving extremely fast. Like anything accelerating too quickly, it will need to find its balance and its place.
That balance will come as people and AI learn to coexist. We’ll get used to it—but humans will also have to up their game. Our thinking, judgment, and adaptability will matter more than ever. It’s exciting and a little scary at the same time, which usually means we’re in the middle of real change.
From a data-driven operator’s view, the signal is clear: advantage won’t come from “better prompts” but from deployment depth. Memory + workflow integration compound like first-party data. Most firms still treat AI as a tool, not an operating layer—leaving massive productivity and decision leverage untapped.
Fantastic breakdown of Altman's interview. The memory-over-raw-intelligence argument really nails why deployment lags so hard behind capability. I've seen this in practice at a couple companies where they keep asking the model the same basic questions, never build context into the workflow, and then complain it's not worth the hype. The toothpaste analogy is spot-on tho, becuz once you experince a model that remembers your preferences and actually learns your patterns, switching back feels broken.
Thanks for this excellent summary and critique of the Altman interview. I find Altman to be a bit of a hype man (understandably), but I do think the business opportunity that no one is really talking about is the effective 'retrofitting' of AI onto legal companies. Someone is poised to make generational fortunes here...
What I’m really waiting for now is AI’s ability to make logical, real-life decisions not just fast or intelligent outputs, but choices that actually make sense in context. Yes, memory is improving and systems are starting to retain more, but AI is still moving extremely fast. Like anything accelerating too quickly, it will need to find its balance and its place.
That balance will come as people and AI learn to coexist. We’ll get used to it—but humans will also have to up their game. Our thinking, judgment, and adaptability will matter more than ever. It’s exciting and a little scary at the same time, which usually means we’re in the middle of real change.
From a data-driven operator’s view, the signal is clear: advantage won’t come from “better prompts” but from deployment depth. Memory + workflow integration compound like first-party data. Most firms still treat AI as a tool, not an operating layer—leaving massive productivity and decision leverage untapped.
Fantastic breakdown of Altman's interview. The memory-over-raw-intelligence argument really nails why deployment lags so hard behind capability. I've seen this in practice at a couple companies where they keep asking the model the same basic questions, never build context into the workflow, and then complain it's not worth the hype. The toothpaste analogy is spot-on tho, becuz once you experince a model that remembers your preferences and actually learns your patterns, switching back feels broken.