AI Briefing#ai-research
The latest AI and developer news about #ai-research, with the original source and a short summary.
The latest AI and developer news about #ai-research, with the original source and a short summary.
Post-training made models more decisive, but it did not improve accuracy.
Researchers from Meta have been moving to Thinking Machines Lab, fueling an ongoing talent war.
TLDR AI is recruiting a remote curator for 3-5 hours a week to pick out the essential AI news.
GPT-5.4 Pro solved Erdős unsolved problem #1196 in 80 minutes.
Anthropic has developed an autonomous AI research agent that surpasses humans, proposing ideas and running experiments.
Running 90 queries across 8 models showed that 86% of results appeared on only one provider.
An experiment interpreting reasoning degradation and catastrophic forgetting as the same kind of structural collapse.
Google Finance has been revamped into a new version equipped with Gemini AI, expanding its service to over 100 countries including Korea.
An OpenAI researcher shared insights on evaluating AI models, post-training, and research methodology.
Bibby AI, a research AI tool that supports everything from literature review to paper drafting and LaTeX error correction, has launched.
OpenAI participated in the First Proof challenge, which verifies AI's mathematical reasoning ability, and confirmed the likelihood of correct answers for 5 or more out of 10 problems.
A support program worth €500,000 is being run to support AI safety and wellbeing for youth in the EMEA region.
Research assistant platform **Consensus** is accelerating the research process through **Scholar Agent**, a multi-agent system built on **GPT-5** and the **Responses API**.
OpenAI has launched the 'OpenAI Learning Accelerator,' combining AI research and education to support educators and students in India.
A researcher at Bain & Company is leveraging OpenAI deep research to analyze complex industry trends and strengthen work capabilities.
OpenAI has secured $6.6B in new investment, reaching a valuation of $157B.
A feature has been integrated that allows machine learning demos to be run instantly from arXiv paper pages via Hugging Face Spaces.
Cohere has launched the **Cohere For AI Scholars Program** for emerging research talent.
Upstage is launching an open recruitment drive for 200 people across all job functions to advance AI Pack.