1,749 mentions · 790 chunks · 131 episodes
...n a given situation, the product doesn't work. In some ways this is reasonable: LLMs are improving the quality-per-unit-cost rapidly. But a better approach is to design a system where the LLM's quality is the floor. If the humans are ...
...r product management is more about behavior. Even with the magical duct tape of LLMs, end-user programming is still hard to achieve. You can get quick, scrappy prototypes quickly, but it's hard to maintain / grow / extend them without...
Instead of handing your agency over to LLMs, use them to extend your agency!
...nd started making bridges in new ways native to it. Similarly, people are using LLMs to make things in more code-like ways, because we don't yet know the best way to work with this new, odd material. The current products are hodgepodg...
A few people have built a way to interact with LLMs where it can ask questions about your day and help you distill a journal for reflection later. One way: do a GPT in ChatGPT. It's easy to do... but n...
LLMs are good at intuition, not reasoning. The "no cake recipe should have 5 tablespoons of tabasco sauce in it" call is not reasoning, but intuition, so ...
LLMs are like a trained circus bear that can make you porridge in your kitchen. It's a miracle that it's able to do it at all, but watch out because no ma...
ChatGPT and LLMs have proven to be more useful to individuals than to companies. That implies that we're still in the community garden phase, not the factory farming ...
For LLMs, we have the camera-making hobbyists today, but not the photographers. At the beginning the skillset has to be camera-making, and over time it shifts...
Using LLMs properly requires LLM-fu. Just like Google-fu back in the day. The kinds of people who had developed an intuition on how to formulate their query (so...
LLMs are not the Thing. They're a part of the Thing. What is the Thing? We don't know yet! LLMs are magical duct tape that can be used to build the Thing....
...tive: lead with a new privacy paradigm allowing composition first, and then add LLMs to that. This could create a system where a copernican shift of data provides leverage for whole new types of experiences to be possible that weren't...
One thing LLMs can help with: making simple judgment calls at the level that any reasonable human could do. "This cake recipe says you should put in five tablespoon...
LLMs are vulnerable to the "screenshot attack". That is, if they say something offensive or wrong the user can take a screenshot and it can go viral, erod...
Stray thoughts on LLMs. LLMs (just like people) are a lot better at critiquing things than coming up with new things. LLMs have a theory of mind, in general. But if they do...
In systems that have a quality component (e.g. search engines, or LLMs), the query stream coevolves with the underlying quality of the service. Users as a population clue into what it can do and give it queries it will d...
A few stray thoughts on LLMs. I love using ChatGPT as a kind of family feud "what will the average (X category of person) think about this phrase". Kind of an automatic wisdom of...
A few riffs on LLMs. An intuition for things that LLMs will get right: if Wikipedia has explained the concepts well. Those facts are likely to also ripple out and inform...
...question?" to ensure you can draft off what happened before. Contrast that with LLMs; they have absorbed a kind of reasoning about the content; a wisdom of the crowds, but also its own kind of emergent wisdom. That means that you can ...
Let's pull on a thread starting from the observation that LLMs only "think" one token at a time. Imagine a prompt like "Write a synopsis of X, and bold the most salient words." The LLM has to choose to emit the m...