Adam Siegel

@thenarrator Except the best traders aren't necessarily the most accurate, they're just...the best traders, so any social currency you assign to them may be misplaced.

the next social network is a prediction market where your feed is ranked by accuracy not engagement. the person who is right 80% of the time gets seen while the person who is loud gets buried

Thanks @marcusabramovi1 for a thought provoking post on funding for forecasting. My thoughts: forum.effectivealtruism.org/posts/zmbfZjK5…

@NathanpmYoung @marcusabramovi1 My own response: forum.effectivealtruism.org/posts/zmbfZjK5…

My comments on the current state of EA and forecasting, contra @marcusabramovi1

@JWerner247 I have often cited you in my own work (completely unrelated to sports) as an example of someone who works extremely hard and does it all “the right way” so know your Dad’s legacy surely lives on through you and those who are impressed by your own drive, talents, and ethics.

I lost my dad today. He was a farmer all his life, and his insane work ethic every day to build us a better life certainly rubbed off on me and my brother. And he was a phenomenal storyteller who loved to chop it up with anyone he met, so maybe he passed that on to me too. Most of all, he was one of the best, most tolerant and most rational men I knew. I’m gonna miss him, but I’m glad we got one last Christmas with him and I’m glad I got to see the joy in his eyes as a grandfather. Tell your loved ones you love ‘em.

We released a new version of Raif, a Rails engine for working with LLMs. It now supports streaming responses in chat & working with provider managed tools like web search/code execution. github.com/CultivateLabs/… It powers all of the LLM features in ARC: arcanalysis.ai

To better understand the implications of the Israeli strikes on Iran, I used ARC to generate a DRAFT of a research brief, outcome scenarios, drivers and indicators to monitor, and a curated list of other perspectives. In about 5 minutes, I had this: docs.google.com/document/d/1uN…

@NunoSempere Went through a similar exercise but used ARC to create an output with an overall research brief, possible scenarios, and a listing of drivers and indicators to be paying attention to (and forecasting!) Maybe it would be helpful input? docs.google.com/document/d/1uN…

Super proud of this!

@0xPetra @robinhanson They were run by Inkling Markets which became @cultivatelabs. They were play money using @robinhanson LMSR.

@robinhanson Do you know which flavor of prediction market they were using? I assume it was a custom implementation.

I'm told the once internal CIA prediction markets are no more, and not because they weren't accurate.

@paulg @robinhanson It was a bureaucratic/funding decision more than efficacy/accuracy. We ran them for 12 years (remember Inkling Markets? :) - project ended in 2020.

@robinhanson Why do you think they got rid of them?

@robinhanson We ran them for over 10 years. They were called the “IC prediction market.” They did get canceled in 2020 but it was more a bureaucratic/funding reason than anything else. They were used across the IC.

Congratulations to @RANDCorporation's Forecasting Initiative (RFI) Pro Forecaster Adjuncts rand.org/global-and-eme…. I've watched them as pro forecasters for a while now: a seriously impressive group and now they're going to help others inside RAND become great forecasters.

@Sn1per_T Daughter is a big @IlliniMBB fan so when we bought these new shoes for basketball she said "These aren't Jason Tatum's, they're Terrence Shannon Jr's. Glad you've been getting some playing time lately!

yalereview.org/article/tavako… Great review, brought back a lot of memories of listening to these guys over and over in my room and on my discman.

🎊Our 2024 wrap-up just hit inboxes! Dive into our CEO @amsiegel's year-end review + get early access to ARC, our new tool combining human expertise with #AI capabilities to help analysts work smarter. Take a look: mailchi.mp/cultivatelabs/…

Prediction markets are having their moment. This time they were directionally correct. Other times, they're not. 2016 headline as a reminder. They absolutely have their utility, but beware the claims of a "better crystal ball." pbs.org/newshour/show/…

Hi @ChipotleTweets my son wanted to let you know your burritos are good enough to dress like one for halloween :)

It's not halloween without @disney 's youtube.com/watch?v=vOGhAV…

Forecasting the Probability of War Between Israel and Hezbollah rand.org/pubs/commentar… via @RANDCorporation

vox.com/future-perfect… Wide-ranging piece about the crowd forecasting space with some nice mentions of our @INFERpub project