Podcast Episode · Possible · 08/13/2025 · 52m

Exploring humanity's vision for artificial intelligence through global conversations and collective intelligence.

Podcast Episode · Possible · 08/13/2025 · 52m

When Large Language Models are used as judges for decision-making across various sensitive domains, they consistently exhibit unpredictable and hidden measurement biases, making their verdicts unrelia...

Exploring humanity's vision for artificial intelligence through global conversations and collective intelligence.

They could offer a more nuanced way to measure AI’s bias and its understanding of the world.

This position paper argues that in many realistic (i.e., complex, contextualized, subjective) scenarios, one LLM is not enough to produce a reliable output. We challenge the status quo of relying sole...

While Large Language Models (LLMs) can exhibit impressive proficiency in isolated, short-term tasks, they often fail to maintain coherent performance over longer time horizons. In this paper, we prese...

Squaring up to the foundational question for language models.

Monitoring the noises within ecosystems reveals their health—allowing researchers to monitor changes in biodiversity, detect threats, and measure the effectiveness of conservation strategies.

Dr. Nelson was an invited speaker at a dinner hosted by French President Emmanuel Macron at the Palais de l'Élysée on February 10, 2025.

Reimagining the technical machine learning benchmarks that drive model development to reflect and encode public values.

Exploring humanity's vision for artificial intelligence through global conversations and collective intelligence.

Arc Institute develops the largest AI model for biology to date in collaboration with NVIDIA, bringing together Stanford University, UC Berkeley, and UC San Francisco researchers

The rapid development of advanced AI agents and the imminent deployment of many instances of these agents will give rise to multi-agent systems of unprecedented complexity. These systems pose novel an...

This paper argues that market governance mechanisms should be considered a key approach in the governance of artificial intelligence (AI), alongside traditional regulatory frameworks. While current go...

To enable artificial agents to generate human-like goals, a model must capture the complexity and diversity of human goals. Davidson et al. model playful goals from a naturalistic experiment as reward...

Recent discussions and research in AI safety have increasingly emphasized the deep connection between AI safety and existential risk from advanced AI systems, suggesting that work on AI safety necessa...

In a new paper, digital activist Audrey Tang and others emphasize the need for context and community over clicks.

The Model Spec specifies desired behavior for the models underlying OpenAI's products (including our APIs).

Scaling inference compute enhances reasoning in large language models (LLMs), with long chains-of-thought (CoTs) enabling strategies like backtracking and error correction. Reinforcement learning (RL)...

Large language models (LLMs) are helping millions of users write texts about diverse issues, and in doing so expose users to different ideas and perspectives. This creates concerns about issue bias, w...

Southeast Asia’s developers have sought to democratize AI by building language models that better represent the region’s languages, worldviews, and values. Yet, language is deeply political in a regio...

Announcing our new Economic Index, and the first results on AI use in the economy