Recently, an AI model compliance checking tool developed by the Swiss startup LatticeFlow has drawn widespread attention. The tool tested generative AI models developed by several major tech companies, including Meta and OpenAI, revealing significant shortcomings in key areas such as cybersecurity and biased outputs. Image source note: The image was generated by AI, licensed by image service provider Midjourney since OpenAI released ChatGPT at the end of 2022.

Recently, new data released by payment technology company Stripe has garnered widespread attention, showing that artificial intelligence (AI) startups are exhibiting an extraordinary speed in revenue growth. Unlike traditional software companies, these AI firms can reach millions of dollars in sales within just a year, which has surprised many. Note: Image generated by AI, licensed by Midjourney. In previous tech startup waves, many startups took years to achieve significant profitability, but now...

OpenAI and several tech companies oppose a California bill aimed at ensuring the safety and transparency of artificial intelligence, fearing that it will hinder innovation and development in AI technology, potentially leading to an outflow of the industry. OpenAI stated in a press release that overly strict regulations could stifle the creativity of developers and researchers, limiting exploration and experimentation. They emphasized that technological innovation requires flexibility rather than being constrained by legal limitations. OpenAI hopes to engage in dialogue with lawmakers to promote a more reasonable legal framework that aligns with industry development trends. If the bill passes, it could lead to...

Australia has released new online safety standards, causing concerns among tech companies. Microsoft, Meta, and others worry that the standards may weaken AI system protections. The new standards target generative artificial intelligence, addressing child safety and terrorism-related content. Tech companies believe the standards could limit AI training and affect the accuracy of content moderation.

The House of Lords committee stated that tech companies are infringing creator rights by using content without permission. The committee urged the government to address copyright issues and called for action. The development of generative AI has highlighted the urgency of copyright matters, and the government should provide more protection. Creators have claimed their works are being illegally used to train large language models, and the committee suggested that the government update the legislative framework.

Cai Chongxin pointed out at the 2023 Yunqi Conference that the era of intelligence has begun, and AI will become a new form of productivity across various industries. 80% of Chinese tech companies and half of the large model companies rely on Alibaba Cloud services. The AI open-source community 'Modao' launched by Alibaba attracted 2.7 million developers in one year, with over 2,300 high-quality models.

MIT professor Max Tegmark has issued a warning that AI tech companies are caught in an out of control race. He calls for a six-month pause on the development of large AI systems, but this has not been successful. Tegmark warns that AI development could spiral out of control, urging government intervention. He emphasizes that models should not be enhanced, nor should dangerous technologies be open-sourced until safety standards are met. Concerns have been raised about the development of AI, including issues related to deepfake videos and potential threats.

New Australian regulations require tech giants to eliminate AI-generated child abuse images. The regulations mandate tech companies to regularly review and improve AI tools to address the issue of child abuse images potentially generated by artificial intelligence. The eSafety Commissioner urges tech companies to reduce the harms caused by their products, requiring search engines to remove child abuse material and prevent the generation of deepfake images.