98% sounds like a lot. If someone wins the lottery 98% of the times they play, they are clearly blessed. Getting a top mark (e.g.: 10/10) on exams 98% of the time will likely lead to an honour diploma.
Re-evaluating GPT-4’s bar exam performance
Perhaps the most widely touted of GPT-4’s at-launch, zero-shot capabilities has been its reported 90th-percentile performance on the Uniform Bar Exam. This paper begins by investigating the methodological challenges in documenting and verifying the 90th-percentile claim, presenting four sets of findings that indicate that OpenAI’s estimates of GPT-4’s UBE percentile are overinflated. First, although GPT-4’s UBE score nears the 90th percentile when examining approximate conversions from February administrations of the Illinois Bar Exam, these estimates are heavily skewed towards repeat test-takers who failed the July administration and score significantly lower than the general test-taking population. Second, data from a recent July administration of the same exam suggests GPT-4’s overall UBE percentile was below the 69th percentile, and $$\sim$$48th percentile on essays. Third, examining official NCBE data and using several conservative statistical assumptions, GPT-4’s performance against first-time test takers is estimated to be $$\sim$$62nd percentile, including $$\sim$$42nd percentile on essays. Fourth, when examining only those who passed the exam (i.e. licensed or license-pending attorneys), GPT-4’s performance is estimated to drop to $$\sim$$48th percentile overall, and $$\sim$$15th percentile on essays. In addition to investigating the validity of the percentile claim, the paper also investigates the validity of GPT-4’s reported scaled UBE score of 298. The paper successfully replicates the MBE score, but highlights several methodological issues in the grading of the MPT + MEE components of the exam, which call into question the validity of the reported essay score. Finally, the paper investigates the effect of different hyperparameter combinations on GPT-4’s MBE performance, finding no significant effect of adjusting temperature settings, and a significant effect of few-shot chain-of-thought prompting over basic zero-shot prompting. Taken together, these findings carry timely insights for the desirability and feasibility of outsourcing legally relevant tasks to AI models, as well as for the importance for AI developers to implement rigorous and transparent capabilities evaluations to help secure safe and trustworthy AI.
Sᴜɴʀɪsᴇ Oᴀᴛʜ 十 Tʜᴇ Sʏʟʟᴀʙᴜꜱ (ORI) - CC0 on Twitter / X
Grades are fine when they are clearly treated as one of many games. The issue comes from worshipping the school and credential system, and so acting like getting 100 on a test or a project is equivalent to entering heaven.— Sᴜɴʀɪsᴇ Oᴀᴛʜ 十 Tʜᴇ Sʏʟʟᴀʙᴜꜱ (ORI) - CC0 (@sunriseoath) October 28, 2026
9Axes
9Axes, based off of 8values is a political quiz that attempts to assign percentages on nine different political axes. You will be presented by a statement, and then you will answer with your opinion on the statement, from Strongly Agree to Strongly Disagree, with each answer slightly affecting your scores. At the end of the quiz, your answers will be compared to the maximum possible for each value, thus giving you a percentage. Answer honestly!
The Role of Luck in Life Success Is Far Greater Than We Realized
Are the most successful people in society just the luckiest people?
Prospect Theory, Mental Accounting, and Differences in Aggregated and Segregated Evaluation of Lottery Portfolios
If individuals have to evaluate a sequence of lotteries, their judgment is influenced by the presentation mode. Experimental studies have found significantly higher acceptance rates for a sequence of lotteries if the overall distribution was displayed instead of the set of lotteries itself. Mental accounting and loss aversion provide an easy and intuitive explanation for this phenomenon. In this paper we offer an explanation that incorporates further evaluation concepts of Prospect Theory. Our formal analysis of the difference in aggregated and segregated portfolio evaluation demonstrates that the higher attractiveness of the aggregated presentation mode is not a general phenomenon (as suggested in the literature) but depends on specific parameters of the lotteries. The theoretical findings are supported by an experimental study. In contrast to the existing evidence and in line with our theoretical results, we find for specific types of lotteries an even lower acceptance rate if the overall distribution is displayed.
Grading is a Scam (and Motivation is a Myth) | A Professor Explains
Unofficial Results - 2026 Referendum
Reporting percentage is calculated based on Voting Area/Locations activity. A Voting Area/Location is considered "reported" once at least one question for that Voting Area/Location has been submitted. A 100% completion rate indicates that every Voting Area/Location has started, but does not necessarily mean every individual question for those Voting Area/Locations has been completed.
Childhood and Education #21: Grades and Standards
First you get to college via the wrong metrics, as previously discussed in #12. Then things get easier.
Access to Opportunity in the Sciences: Evidence from the Nobel Laureates
Unequal opportunity in the sciences reduces scientific contributions from the most talented individuals and limits the rate of human progress. We study unequal opportunity by collecting data on the childhood SES of Nobel laureates in the sciences. The average laureate grew up in an 87–90th percentile household. Access to opportunity doubled from 1901–2023, but remains highly unequal. Barriers are higher for women, but lower for Americans. Access to opportunity across countries is much less equal, and has barely improved at all. Cities with more intergenerational mobility produce more laureates from non-elite families, and more laureates overall.
A belief in meritocracy is not only false: it’s bad for you | Aeon Ideas
The belief that hard work and talent make for success is not only false, it encourages selfishness and discrimination too
Privilege's upward-facing window.
Somewhat recently there was a controversial Twitter thread that defined which universities had computer science graduates worth hiring, and the hirable percentage from those universities. The best ranked universities were small, famous and expensive. Quite a few folks reacted against this perspective’s bias and privilege, at least as many chimed in to confirm the claims.
Selection into Financial Education and Effects on Portfolio Choice
We study the impact of a financial education program on portfolio choices in a setting that allows us to examine both participation decisions and differences in program benefits across individuals. We directly elicit willingness to pay for financial education and show that individuals with higher financial literacy, greater confidence in applying new knowledge, and higher expected returns are willing to pay more. Using incentivized portfolio allocation tasks, we develop outcome measures tailored to individual risk preferences and show that financial education improves decision quality along these dimensions. The program increases the share of participants who experience an improvement by 20 percentage points, corresponding to average gains of 3.2 percentage points of initial wealth. Individuals with the highest willingness to participate benefit disproportionately from the program, with important implications for the design and evaluation of financial education initiatives. By isolating a specific investment skill in a setting with voluntary participation, our design sheds light on how selection into financial education shapes measured program effects and informs the assessment of broader workplace and public training programs. This paper was accepted by Marie Claire Villeval, behavioral economics and decision analysis. Funding: The authors thank the Pension Research Council of Wharton School at the University of Pennsylvania, as well as the Retirement and Savings Institute at HEC Montréal, for generous research funding. Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2024.08268 .
You're Going to Live Past 90. Congrats! Here’s How to Pay For It.
Advances in medicine mean that reaching your nineties and beyond is more realistic than ever. Start preparing now to avoid running out of money.
No One Dies in this Dystopia
I am all done with grading for the semester just past, but the work sticks to me in a way I can’t yet manage to fling off.
Why students are throwing tons of money at a program that won’t give them a college degree
There's a hidden dark side of one of the hottest trends in higher education.
