Search AI Products and News

Mianbi Intelligence announced the completion of a new round of tens of millions yuan financing, which was jointly invested by Hongtai Fund, Guozhong Capital, Qingchuan Golden信 and Maotai Fund. The completion of this round of financing will lay a solid foundation for Mianbi Intelligence in areas such as efficient large model technologies, product barriers, industry empowerment, and ecological expansion, promoting the scaling application of the "edge-side brain" in thousands of industries.

The non-profit organization Digital Promise recently announced the launch of its "Responsible Design AI Product Certification" program. This initiative aims to assist school leaders in selecting AI-powered educational technology tools that meet student learning and safety requirements. With the rapid growth of AI in education, schools face an increasing number of choices, making the selection of safe and effective tools a critical challenge. The certification program specifically assesses several key aspects of AI educational tools, including data security and fairness.

AI company Anthropic recently updated its "responsible scaling" policy, outlining which models require additional safety protections. This move aims to mitigate potential risks before new technologies are released. According to Anthropic's blog, if stress testing reveals an AI model could assist a "resource-constrained state actor" in developing chemical and biological weapons, Anthropic will delay the technology's public launch.

Recently, Google removed content from its official website that previously committed to not developing artificial intelligence (AI) for weapons or surveillance, sparking widespread discussion and concern. According to a Bloomberg report, this change occurred when Google updated its public AI principles page, and the section that explicitly mentioned 'applications we will not pursue' was completely deleted, which came as a surprise. Image Source Note: The image is AI-generated, and when asked for comments on this change, Google referred to the technology.

At a recent forum held in Davos, Deloitte released a report revealing the difficulties companies face in translating generative artificial intelligence (Gen AI) experiments into practical applications. Although companies have achieved positive initial results in generative AI projects, there are still many obstacles to scaling deployment. Source of the image note: Image generated by AI, image licensed by service provider Midjourney. According to the report, many companies find that the transformation process from the lab to production environments is not easy, especially.

A research team from New York University, MIT, and Google has recently proposed an innovative framework aimed at addressing the bottleneck of inference time scaling in diffusion models. This groundbreaking research transcends the traditional method of simply increasing the denoising steps, opening up new avenues for improving generative model performance. The framework primarily unfolds from two dimensions: one is leveraging feedback from validators, and the second is implementing algorithms to discover better noise candidates. The research team builds on a pre-trained SiT-XL model with a resolution of 256×256, maintaining 250 fixed denoising steps.

In today's rapid development of Large Language Models (LLM), the training and inference costs of models have increasingly become a focus of research and application. Recently, the Tencent Hunyuan team released an important study that delves into the 'Scaling Laws' of low-bit floating point quantization training, which refers to the principles governing the scale of floating point quantization training. The core of this research lies in exploring how to significantly reduce computational and storage costs without sacrificing performance by lowering the precision of the model.

According to media reports, Tan Xu, former Chief Research Manager at Microsoft Research Asia, joined the startup Dark Side of the Moon in August, primarily responsible for developing end-to-end speech models. One important goal after Tan's joining is to create a speech experience similar to GPT-4o for the company. Dark Side of the Moon has not yet responded, but insiders have confirmed Tan's employment, and specific position details have not been finalized. Public information indicates that Tan Xu's research areas include deep learning, language/speech/music processing, and AI content generation, with publications in AI academic conferences.

Apple released a technical paper detailing the generative AI models developed for its 'Apple Intelligence' series, set to be integrated into iOS, macOS, and iPadOS. The company addressed ethical concerns by stating that it does not use private user data, instead relying on public and licensed data for training. The paper reveals the sources of data for Apple's 'Apple Foundation Model' (AFM), including public web data, publisher-licensed data, and....

Apple has joined the U.S. government's voluntary AI safety commitment, seen as a key step toward promoting responsible innovation in AI. The commitment announced by the White House includes transparency about AI capabilities and labeling AI-generated content, aimed at reducing misinformation and safety risks associated with advancements in AI technology. Federal agencies have completed actions related to AI safety, focusing on risk management, attracting talent, and enhancing global leadership in AI. Apple's new AI features are facing delays, originally planned for release in the September system update, now set to be postponed by at least a few weeks. Developers will have an early experience with Apple's new A.

Stability AI recently released a more lenient license for Stable Diffusion3 amidst controversy, but is it enough to resolve the issues? Stability AI introduced restrictive new license terms last month, sparking opposition from Stable Diffusion users. However, the company announced on Friday evening that it had relaxed the conditions, receiving mixed reactions from users.Stability AI released an improved community license for the Stable Diffusion3 (SD3) model, aiming to calm the controversy that

OpenAI and DeepMind have differing views and approaches in the research of Scaling Laws. Scaling Laws can predict changes in loss when varying model parameters, data amounts, and computation resources. This competition will drive the development of artificial intelligence and shape the future of human-machine coexistence. Pre-training large language models involves trade-offs between model size, data quantity, and training costs. Scaling Laws can assist in optimizing design decisions. DeepMind argues that model size and data quantity should scale equally.

The UK government's response to AI regulation consultations reveals its regulatory plans in the field of artificial intelligence. The Department for Science, Innovation and Technology promotes a reliance on existing laws and regulatory bodies while also providing over £100 million in funding to strengthen AI regulation. The UK government will invest over £90 million to establish nine research centers aimed at promoting AI innovation in areas including healthcare, mathematics, and chemistry. The government insists on not introducing new AI regulations but will continue to rely on existing laws and empower regulatory bodies to address issues more effectively.

Google will participate in the Global Partnership on AI (GPAI) summit, committing to make machine learning and artificial intelligence more beneficial for users. The company will take steps to address synthetic media, including labeling modified or synthesized content on Search and YouTube. Google has invested $1 million to establish a multidisciplinary Responsible AI Center at the Indian Institute of Technology Madras. Google is dedicated to the development of responsible artificial intelligence and is taking action to combat the deepfake challenge.