AI researchers seek new tools that could slow automated model development

A group of today’s technology workers published Open letter calling for a new approach to regulating automated artificial intelligence development initiatives.

All 1,134 signatories to this document work for companies building cutting-edge AI models. Industry players listed include Anthropic PBC, OpenAI Group PBC, Google LLC, and Meta Platforms Inc.. The letter was also signed by staff from venture-backed AI laboratories such as Thinking Machines Inc. and London-based Inherent Laboratories Ltd.

Supporters of this effort include some of the most prominent figures in the AI ​​field. Anthropic CEO Dario Dario Amodei signed the letter along with several of the company’s other co-founders. They were joined by lead scientists from OpenAI, Meta’s Meta AI unit, and Thinking Machines.

The letter focuses on efforts by large tech companies to use AI to automate model development programs. The signatories are concerned that advances in this field may pose new safety risks.

“It is difficult to predict exactly how much this will accelerate advances in AI. However, there is a real risk that capacity development will accelerate so rapidly that it will outstrip our ability to understand or control the resulting systems,” the letter reads.

The document goes on to argue that the industry needs a way to “deliberately coordinate” automated AI development efforts. According to the signatories, the tools needed to achieve research stagnation are not currently available. The letter asks the U.S. government for support in developing the necessary technology.

“We call on the U.S. government to support international efforts to develop the necessary technical and governance tools to intentionally advance the frontiers of automated AI development,” the signatories wrote.

OpenAI in April revealed Its GPT-5.5 large language model was said to have helped optimize the cluster of Nvidia Corp. graphics cards on which it runs. One of the optimizations developed by LLM has increased the token generation speed by more than 20%. OpenAI has since released a new model lineup that is even better at coding tasks: the GPT-5.6 series.

Anthropic in April too disclosed It is said that Claude is being used to accelerate AI safety research. The company has developed an AI agent powered by Claude to study a machine learning technique called weak-to-strong supervision. Anthropic researchers believe this technology could hold the key to effectively monitoring future AI models.

Today’s open letter continues a recent trend in which researchers are calling for changes in AI policy.

On Friday, Nvidia and other major technology companies prompted U.S. policymakers will not ban open source AI models. A few days ago, a group of economists issued A statement calling for stronger AI regulations.

Google DeepMind CEO Demis Hassabis also joined the fray earlier this month. substack essay. Like today’s open letter, this document suggests that the U.S. government will support the launch of new initiatives to address AI safety risks. Hassabis proposes the creation of a standards body that would develop AI risk benchmarks and use them to evaluate new LLMs before they are widely released.

Image: Pixabay