@yshemesh

PhD student at @ BGU, researching the psychology of prediction using ML | NLP Co-author of How We https://xitter.cf/t.co/KrYpC0uBY2, and Why it Matters.

 Israel
Joined July 2012
  • Tweets888
  • Following246
  • Followers195
29Media
Yhonatan Shemesh retweeted
Victor Kumar@victorckumarJul 15
Fascinating! When participants are presented with arguments involving fMRI data, they showed less activity in an area of the brain associated with rational reflection.
3
16
1
85
5,072
Looking forward to reading this. If it is as good as his previous books (Moral Animal, Non-zero, Why Buddhism is True), this will be a treat.
Robert Wright
@robertwrighterJun 23
Happy pub date to me! My book on AI is now available. If you want to ask questions about it before buying, just click reply. (And feel free to retweet!) The link to Amazon is in my bio.
1
1
1,098
Brexit is a rare forecasting laboratory. Few political decisions generate so many explicit predictions, so clearly dated, and so easily compared with reality a decade later. The results are revealing. 1/5 pif.bgu.ac.il/wp-content/upl…
1
1
4
694
Yhonatan Shemesh retweeted
Michael Gilead
@Michael_GileadMar 25
📢New preprint alert📢 When Crowds Fail: Predicting Failures in Collective Wisdom through Discourse Cues We used psychologically-informed feature extraction from texts to predict wisdom of the crowds in geopolitical forecasting (using @metaculus forecasts). In a pre-registered analysis on a hold-out, our model was able to predict Wisdom of the Crowds failures pretty well (r = 0.535). This cool work was led by Maayan Pe'er, see link in first comment.
1
1
6
474
Yhonatan Shemesh retweeted
Dan Williams
@danwilliamsphilMar 23
Excellent essay - recommend
Eric Levitz
@EricLevitzMar 23
Social media undermined expert authority, democratized public debate, and steered individuals into evermore bespoke conceptions of reality. But AI could reverse these trends. As @danwilliamsphil and @dylanmatt argue, LLMs appear to be a "converging" and "technocratizing" technology -- one that increases expert influence and social consensus. This is partly because AI companies and social media firms have radically different business models: vox.com/technology/483455/ai…
1
10
1
42
10,357
דעה שכתבתי היום ב ΤheMarker. בעידן של קיטוב פוליטי, דיסאינפורמציה וחוסר אמון בממסד, שוקי החיזוי מחזירים את האמת לאופנה. themarker.com/wallstreet/202…

מי מפחד מפולימרקט? שוקי חיזוי יכולים לרפא את עידן הפוסט־אמת

***

themarker.com
23
Yhonatan Shemesh retweeted
Steven Pinker
@sapinkerMar 11
The God Debate: Ross Douthat v. Steven Pinker
The Free Press
@TheFPMar 11
Can a post-God society survive? THE GOD DEBATE is out now on The Free Press and Paramount+. @SAPinker faces off against @DouthatNYT, moderated by @WhigNewtons. Watch it now: thefp.com/p/debate-do-we-nee…
26
2
180
53,921
Yhonatan Shemesh retweeted
Dan Williams
@danwilliamsphilMar 4
New essay! When it comes to thinking about the impact of LLMs like ChatGPT, Claude, Gemini, and even Grok on public opinion, I argue that they function as a kind of anti-social media. Whereas social media has been democratising, epistemically diverging, engagement-optimised, and performative, LLMs are "technocratising", epistemically converging, accuracy-optimised, and polite. conspicuouscognition.com/p/h…
27
117
16
486
75,811
The Brier score is the standard metric for evaluating forecasting accuracy. But it has a major drawback: it’s difficult to interpret and communicate. For example, superforecasters have a Brier score of 0.086 on ForecastBench, our AI forecasting benchmark. Technically, this means that on average superforecasters are “0.086 squared probability units from the truth.” Confused? You’re not alone. We propose a simple fix: the Brier Index. 🧵
2
5
32
2,260
תחרות הניבוי השנייה של מבט-על - "מבט לסוף השנה" - יצאה לדרך! חושבים שאתם יודעים מה צופן העתיד? ענו על שאלות, נמקו - ואולי תזכו בפרסים! התחרות פתוחה עד יום שישי 07/11. פרסים יינתנו גם על הסברים מנומקים בהזנת התחזיות ובתגובות בפורום. פרטים והרשמה: app---98d8ae29.base44.app
1
2
219
Yhonatan Shemesh retweeted
Yonatan Kaplan@Yonatan_Kaplan_3 Nov 2025
האם ניתן לחזות את העתיד באמצעות בינה מלאכותית? בהנחיית @Michael_Gilead, פיתחתי יחד עם @yshemesh תשתית שמאפשרת למערכות AI לדמות תהליכי חשיבה של מומחים, וכך לחזות תרחישים גאופוליטיים. haaretz.co.il/magazine/2025-…

זה לא דמיוני כמו שזה נשמע: אנחנו מתקרבים לרגע שבו נוכל לחזות את העתיד

הארץ

haaretz.co.il
1
1
5
286
Yhonatan Shemesh retweeted
Michael Gilead
@Michael_Gilead31 Oct 2025
כתבה השער של מוסף סופ"ש של הארץ עוסקת בעולם הניבוי הגיאופוליטי ובמחקרים שלנו (@Professologue, @yshemesh, מעיין פאר, יהונתן קפלן, הילה שיניצקי). הכתבה של גדעון לב מאוד מעניינת באופן כללי, ומדווחת על תמורה היסטורית משמעותית שאנחנו נמצאים במהלכה. haaretz.co.il/magazine/2025-…
3
5
27
1,209
New paper out 📢 Can we tell which forecasts to trust? Studying superforecasters and platforms like @metaculus and @Polymarket shows that accurately predicting the future is achievable through a combo of good cognitive practices and proper incentives. The problem? 1/8
1
2
1
11
1,170
more replies
The results reveal a gap between perceived and actual forecasting competence. They show how social instincts (=following charismatic leaders) can undermine our pursuit of accurate beliefs, and highlight the need for accountability systems essential for learning whom to trust. 7/8
1
66
Authors: @Michael_Gilead, @Professologue, and Avi Gammoran. We thank @metaculus and @tamaybes for their collaboration. We thank @mattansb for consulting on statistical methods and Hilla Shinitzky for her involvement in the development of the study. 8/8
3
109
Load more