Next Newsletter

Explore the Index of Economic Freedom to gauge global impacts of liberty and free markets. Discover the powerful link between economic freedom and progress. The 30th edition, once again, illustrates key factors shaping our world's landscape. From @Heritage

In Gaza, hunger is taking a toll on the bodies of children. The impact can last a lifetime.

We analyzed more than 4.1 million U.S. immigration court records amid debates about the U.S.-Mexico border, which often omit what happens to migrants who stay.

We built a tool to reveal the channels used by AI giants

Everyone complains about the weather forecast, but some people have more cause to grumble than others. Find out how accurate the forecast is where you live.

Make sense of emotions with this interactive feelings wheel. Robert Plutchik's wheel of emotions explained - guide & free download

Explore how will your city feel in the future.

How do our comedic sensibilities form and transform over time?

Tic-Tac-Toe reimagined as a game of chance.

A data-driven essay identifying patterns in how I fill out the sudoku.

Over 1000 surveys and other data collections related to American and international religion are available for online preview and download

The release of the first wave of a 200,000-person study on flourishing.

To deepen the public conversation about how AI models should behave, we’re sharing the Model Spec, our approach to shaping desired model behavior.

Terminal can be quite fun to play around with. On encouragement of my friend Pablo, I ventured around to discover some interesting ones. To use any of them, you should fire up Terminal on your MacOS/Linux/Unix and install the right packages using apt-get on Linux or brew on MacOS. First, install Homebrew if you don’t have it already. /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)" Once you have that, install any given package using brew install command like below.

The development and popularization of large language models (LLMs) have raised concerns that they will be used to create tailor-made, convincing arguments to push false or misleading narratives online. Early work has found that language models can generate content perceived as at least on par and often more persuasive than human-written messages. However, there is still limited knowledge about LLMs' persuasive capabilities in direct conversations with human counterparts and how personalization can improve their performance. In this pre-registered study, we analyze the effect of AI-driven persuasion in a controlled, harmless setting. We create a web-based platform where participants engage in short, multiple-round debates with a live opponent. Each participant is randomly assigned to one of four treatment conditions, corresponding to a two-by-two factorial design: (1) Games are either played between two humans or between a human and an LLM; (2) Personalization might or might not be enabled, granting one of the two players access to basic sociodemographic information about their opponent. We found that participants who debated GPT-4 with access to their personal information had 81.7% (p 0.01; N=820 unique participants) higher odds of increased agreement with their opponents compared to participants who debated humans. Without personalization, GPT-4 still outperforms humans, but the effect is lower and statistically non-significant (p=0.31). Overall, our results suggest that concerns around personalization are meaningful and have important implications for the governance of social media and the design of new online environments.

When does our taste in music stagnate?

Search engine and newsletters for developers and tech professionals. Keep up to date with tech news, blogs and research from over 30,000 sources with personalised, AI-curated newsletters.

Build AI Assistants with memory, knowledge and tools. - phidatahq/phidata

Watch hundreds of teenagers grow up into adults – and see how their lives turn out

Two-thirds of adults get at least 7 hours of sleep. I am not in that two-thirds.

Within this video, a deep-fake of Ryan Gosling (using ElevenLabs for the voice and SyncLAbs for the lip syncing) give a high-level explanation how LLMs work.

Are LLMs good enough yet to solve the business and societal challenges we are setting them up for?

How one crew risked radiation, storms, and currents to save Japan from digital isolation.