China's Zhipu AI releases GLM-5.2, rivals US models in cybersecurity tasks

·1 views

Chinese AI firm Zhipu AI has released its open-weight model GLM-5.2, which some researchers say performs comparably to Anthropic's Mythos in bug-finding and cybersecurity applications. The model still trails behind leading US models from Anthropic and OpenAI in general-purpose tasks. The development signals a significant narrowing of the AI capability gap between China and the United States. This progress has raised concerns within the US government, which has been actively restricting China's access to advanced AI models and the hardware required to train them. The Trump administration has treated models like Mythos as strategically sensitive technologies in the context of US-China competition.

Read the full story at The Verge

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

California law targeting loud streaming ads takes effect on July 1

Streaming ads might be getting a lot quieter.

0 comments Read more at TechCrunch

TechnologyThe Verge ·

Suno Launches Artist Incubator Offering Grants, Mentorship, and Broad Music Licenses

AI music platform Suno has introduced Spark, an incubator program aimed at supporting independent artists through grants, mentorship, and marketing resources. The program is open to unsigned singers, songwriters, and producers who release music under their own name. Applicants must agree to terms that include making their songs available on Suno for remixing by other users. The licensing terms have drawn scrutiny from users on the Suno subreddit, who raised concerns about the breadth of rights granted to the company over participants' work. The initiative signals Suno's broader ambition to evolve beyond an AI music generator into a streaming and artist discovery platform.

0 comments Read more at The Verge

TechnologyTechCrunch ·

Ford rehires veteran engineers after AI fails to meet quality standards

Ford has rehired experienced senior engineers, internally referred to as 'gray beards', after the company found that artificial intelligence alone could not ensure high-quality vehicle production. The automaker had previously reduced its reliance on seasoned human expertise, expecting AI tools to fill the gap. However, Ford acknowledged the assumption was a mistake, as AI fell short of delivering the expected product quality. The company's reversal highlights the continued importance of human experience and judgment in complex manufacturing processes.

0 comments Read more at TechCrunch

TechnologyArs Technica ·

Academic Journal Retracts Two 1940s Papers by Nobel Physicist Max Planck

A scientific journal has retracted two papers originally published in the 1940s by renowned physicist Max Planck, the Nobel Prize-winning father of quantum theory. The retracted articles are no longer accessible, with their links now leading to blank pages and empty PDF files. The move has drawn criticism from academics, with at least one expert calling the decision intellectually unacceptable. The retraction of decades-old historical scientific work raises broader questions about the standards and ethics of removing legacy publications from the scholarly record.

0 comments Read more at Ars Technica