News list for "study"

An emerging class of AI systems that rewrite their own code and workflows may erode their own safeguards, researchers say.

2025-10-01 20:30:37

NetEase-backed study shows language model agents may detect bugs faster and with greater coverage than existing tools.

2025-09-30 02:00:49

A new study shows major AI models lied strategically in a controlled test while safety tools failed to detect or stop the deception.

2025-09-29 20:25:50

A new AI study finds that large language models show stable, surprising behaviors when left alone.

2025-09-26 19:12:58

Updates to DeepMind's Frontier Safety Framework highlight concerns that advanced AI can evade human control and sway user beliefs.

2025-09-22 17:29:55

Researchers showed that AI deep learning models, used in everything from self-driving cars to finance to healthcare, can be sabotaged. Here's how.

2025-08-25 16:51:05

At 1 am today, OpenAI released the ChatGPT Study learning model. This model is mainly for the education field, providing interactive prompts, scaffolding responses, personalized education, and knowledge point inspection. It can deeply interpret the problem-solving ideas and processes of each problem, rather than just giving answers. That is to say, ChatGPT Study is equivalent to a 24-hour online tutor, which can provide personalized guidance and feedback according to students' learning progress ...

2025-07-29 22:01:00

今天凌晨1点，OpenAI发布了ChatGPT Study学习模式。该模式主要面向教育领域，提供交互式提示、支架式回应、个性化教育、知识点检查四种方法，可以深度解读每一道难题的解题思路和流程，而不是仅仅给出答案。也就是说ChatGPT Study相当于一个24小时在线的家教，能够根据学生的学习进度和需求，提供个性化的指导和反馈。不仅能够帮助学生理解复杂...

2025-07-29 22:01:00

Trump: (On international students) still wants international students to continue studying in the United States.

2025-05-30 18:36:59