cl42

11 ms·

121.
Lesser-Known and Publicly Traded AI Companies(emergingtrajectories.com)
1 points by cl42 2y ago|0 comments
122.
by cl42 2y ago
I really dislike this sort of headline because it misses a core part of the analysis between seed-stage, versus late-stage startups. NYT says the $27.1B represents ~50% of investments in startups ($56B in total). But the $27.1B includes $1B
123.
by cl42 2y ago
If you're curious, here's a 165-page Taiwan war game run across multiple scenarios and events: https://www.csis.org/analysis/first-battle-next-war-wargamin...
124.
by cl42 2y ago
I believe in this case that's very much the goal -- less about who wins, and more about the options/tactics debated to inform actual military battle prep.
125.
The Operational Wargame Series: The best game not in stores now (2021)(nodicenoglory.com)
116 points by cl42 2y ago|79 comments
126.
Using LLMs to score and track geopolitical risk in the Sahel region of Africa(emergingtrajectories.com)
1 points by cl42 2y ago|0 comments
127.
by cl42 2y ago
I'm using LLMs to basically build "junior analysts" that monitor very niche types of companies -- think, junior mining companies, or very specific commodities futures... A lot of these spaces have tons of terrible companies
128.
by cl42 2y ago
Thank you! This is so helpful. It's also interesting to see what temperature value they use (1.0, 0.1 in some cases?)... I have a feeling using the actual raw probability estimates (if available) would provide a lot of information with
129.
by cl42 2y ago
1. "Show, don't tell." If you have to explain in detail your value prop and rationally convince a potential user that they will be better off using your product, then you might be wrong about how good the product is. If you&#
130.
by cl42 2y ago
I'm curious how this actually plays out in the long run. Canada is a good example, where the anti-immigration narrative has become much louder very quickly. People get nervous/upset when they see their culture or population visibl
131.
by cl42 2y ago
Do you think it's different when global growth declines, or when there's a large group of countries going throughs population stagnation or decline? One can say that Japan's economy was still buoyed by global population growt
132.
by cl42 2y ago
I was just emailing a few folks in the insurance industry about demographic decline. I can't find any good economic analyses on what will likely happen to our economies when populations start to decline. This will happen given the pr
133.
by cl42 2y ago
I think this is particularly bad in Canada, where most of the brands are owned by two or three companies, and there has been clear evidence (and penalties) for price fixing. Would love to hear what Canadians think about this + if they see a
134.
The Future of AI Is Vertical(insights.euclid.vc)
3 points by cl42 2y ago|0 comments
135.
by cl42 2y ago
Thanks for doing this. I'm a huge pgvector fan and user... After trying a number of different options (Pinecone, ChromaDB, FAISS + memory stores), I felt like pgvector offered the best value and project maturity. Performance was my big
136.
by cl42 2y ago
You're not wrong! After training, the neural network is in fact "hard coded" in that the weights do not change anymore. During training, you update the weights. As a very, very simple example in relation to my original respon
137.
by cl42 2y ago
LLMs are auto-regressive predictors -- so they take the text given to them (i.e., the prompt) and generate a probability estimate for what the next token should be. Suppose you gave it a quote -- "Once upon a midnight dreary, while I p
138.
by cl42 2y ago
The NYT has a great podcast on this[1] -- insurers are refusing to insure geographic areas that are seeing a spike in weather like this. Not sure if this applies to car insurance, but home insurance is definitely changing. 1: https:/&
139.
A List of 160 Data Centers(emergingtrajectories.com)
2 points by cl42 2y ago|0 comments
140.
by cl42 2y ago
Yeah, I feel you sometimes need to provide the actual verbatim quote the LLM is referencing for the fact, otherwise it's unclear how the LLM 'interpreted' the information for you.
141.
by cl42 2y ago
You can 100% do this. We do that at my startup ( https://emergingtrajectories.com/ ). We build a fact base from reliable sources or information you give us, and we have multiple layers of checks to ensure the LLM only referen
142.
by cl42 2y ago
Hard to fully predict the future, of course, but changes are the answer is "no". Assuming you are comfortable with the current model performance rather than always trying to have the most performant model... 1. Lots of researchers
143.
Rocket Lab – space industry dynamics(emergingtrajectories.com)
4 points by cl42 2y ago|0 comments
144.
by cl42 2y ago
> Although all five Europa images from Juno are high-resolution, the image from the spacecraft’s black-and-white SRU offers the most detail. Designed to detect dim stars for navigation purposes, the SRU is sensitive to low light. To avoi
145.
by cl42 2y ago
This is a beautiful answer. You're right to point out the need for social connection, entertainment, play, etc. and how these are all critical aspects of human life. "We'll always find ways to entertain ourselves." Your
146.
by cl42 2y ago
Interesting! I wonder if this will be the way we open the door to microtransactions on the web!
147.
by cl42 2y ago
Do you think we'll return to human-curated directories?
148.
Ask HN: What are your predictions for the Internet in 2-5 years?
22 points by cl42 2y ago|29 comments
149.
by cl42 2y ago
Life-changing event/experience. For those debating driving out to a dark area the next time this happens -- just go to a local park or part of your neighborhood where trees or other obstacles block the light. You need not drive 2 hours
150.
by cl42 2y ago
Side note / point: I'm a current WeWork customer and have been incredibly impressed with the consistency of their offering. Every location is the same (in a good way)... Even down to the coffee, art, phone booths, etc. Has anyone

More ›