Responsible AI

Google revised its AI principles, reversing previous commitments to avoid work on weapons, surveillance, and other military applications beyond non-lethal uses like communications, logistics, and medicine.

Anthropic analyzed 1 million anonymized conversations between users and Claude 3.5 Sonnet. The study found that most people used the model for software development and also revealed malfunctions and jailbreaks.

As we approach 2025, my greatest hope for AI is that it will enable prosocial platforms that promote empathy, understanding, and collaboration rather than division.

Building a foundation model takes tremendous amounts of data. In the coming year, I hope we’ll enable models to learn more from less data.

Large language models that remember more hallucinate less.

Jailbreak prompts can prod a large language model (LLM) to overstep built-in boundaries, leading it to do things like respond to queries it was trained to refuse to answer. Researchers devised a way to further boost the probability that LLMs will respond in ways that respect such limits.

Two top AI companies changed their stances on military and intelligence applications.

A new study suggests that leading AI models may meet the requirements of the European Union’s AI Act in some areas, but probably not in others.

Major AI companies plan to meet the growing demand with nuclear energy.

Seemingly an innocuous form of expression, ASCII art opens a new vector for jailbreak attacks on large language models (LLMs), enabling them to generate outputs that their developers tuned them to avoid producing.

Large language models can produce output that’s convincing but false. Researchers proposed a way to identify such hallucinations. 

At meetings in Seoul, government and corporate officials from dozens of countries agreed to take action on AI safety.

The United States and China opened a dialogue to avert hypothetical AI catastrophes.

OpenAI models were used in five disinformation campaigns, the company said.

OpenAI pulled back the curtain on revised rules that will guide its models.