AI
We can also imagine interfaces that support iterative shaping: “More like this, less like that.” The commonality is that these systems take as a given that initial prompts are often just the starting point for a shared search process.
We don’t need humans to become hyper-specific, up-front planners. We need systems that meet us where we are: in the ... See more
We don’t need humans to become hyper-specific, up-front planners. We need systems that meet us where we are: in the ... See more
🌀🗞 The FLUX Review, Ep. 190
In 1854, London was gripped by a deadly cholera outbreak, and the medical establishment did what it had always done: treat the sick and quarantine the exposed. A classic response built on the assumption that the disease was airborne.
Or in shovel terms, keep digging deeper!
Throw more resources at the known response pattern, without questioning whe... See more
Or in shovel terms, keep digging deeper!
Throw more resources at the known response pattern, without questioning whe... See more
Sangeet Paul Choudary • Don't sell shovels, sell treasure maps
They look like software, but they act like people. And just like people, you can’t just hire someone and pop them on a seat, you have to train them. And create systems around them to make the outputs verifiable.
Which means there’s a pareto frontier of the number of LLM calls you’ll need ot make for verification and the error-rate each LLM introduce... See more
Which means there’s a pareto frontier of the number of LLM calls you’ll need ot make for verification and the error-rate each LLM introduce... See more
Rohit Krishnan • Working with LLMs: A Few Lessons
In a world where knowledge is cheap,
curiosity, curation, and judgment
- signalled well - becomes insanely valuable.
curiosity, curation, and judgment
- signalled well - becomes insanely valuable.
Sangeet Paul Choudary • Humans as 'luxury goods' in the age of AI
geoguessr transcript
Temporary Chat Prompt: Image + You are playing a one-round game of GeoGuessr. Your task: from a single still image, infer the most likely real-world location. Note that unlike in the GeoGuessr game, there is no guarantee that these images are taken somewhere Google's Streetview car can reach: ...
docs.google.com
That’s what Treasure Map AI does. It curates to elevate what matters and exclude what doesn’t. It empowers you with better judgement. And once you see that it works, it leaves you curious for more.
Sangeet Paul Choudary • Don't sell shovels, sell treasure maps
This means you have to add evaluation frameworks, human-in-the-loop processes, designing for graceful failure, using LLMs for probabilistic guidance rather than deterministic answers, or all of the above, and hope they catch most of what you care about, but know things will still slip through.
Working with LLMs: A Few Lessons
LLMs inherently are probabilistic. This is unlike code that we’re used to running before. That’s why using an LLM can be so cool, because they can do different things.
First, economic value requires scarcity of supply.
Air is vital to life. Its intrinsic value is infinite.
But because it’s abundant, it has no economic value in most cases.
Of course, if you’re going Scuba diving, it’s no longer abundant and now commands economic value as compressed air.
Second, economic value requires relevance to demand.
A soldie... See more
Air is vital to life. Its intrinsic value is infinite.
But because it’s abundant, it has no economic value in most cases.
Of course, if you’re going Scuba diving, it’s no longer abundant and now commands economic value as compressed air.
Second, economic value requires relevance to demand.
A soldie... See more