Everyone is welcome to Star us on GitHub:
Distributed full-link causal learning system OpenASCE: https://github.com/Open-All-Scale-Causal-Engine/OpenASCE
Large model-driven knowledge graph OpenSPG: https://github.com/OpenSPG/openspg
Large-scale graph learning system OpenAGL: https://github.com/TuGraph-family/TuGraph-AntGraphLearning
Hot topics in this issue:
Stack Overflow and OpenAI team up to power the world’s most popular large-scale language model
Apple is developing artificial intelligence chips for data centers, and the project is internally codenamed ACDC
Biggest AI funding in UK history: Wayve raises $1.05 billion to bring its Tesla-like self-driving technology to many carmakers
Zhipu AI is developing a domestic Wensheng video model to benchmark Sora, and it will be released as soon as this year.
Microsoft uses MAI-1 to develop the latest model and will compete with Google, Anthropic and OpenAI for the most advanced models
...
01.Stack Overflow and OpenAI collaborate to enhance the world’s most popular large-scale language model
Stack Overflow and OpenAI announced a new API partnership, bringing the two together through Overflow API access to provide OpenAI users with AI tools and the accurate and vetted data foundation they need to quickly find solutions to problems; OpenAI will also draw from Stack Proven technical knowledge from Overflow is rendered directly into ChatGPT, giving users easy access to trusted, accurate, and highly technical knowledge and code powered by the millions of developers who contribute to the Stack Overflow platform 15 years.
Learn more:
https://stackoverflow.co/company/press/archive/openai-partnership
02. Apple is developing artificial intelligence chips for data centers, and the project is internally codenamed ACDC.
Over the past decade, Apple has become a leader in designing chips for iPhones, iPads, Apple Watches and Mac computers. According to people familiar with the matter, the server project is internally code-named Project ACDC (Apple Chips for Data Centers). For Apple's server chips, this component may focus on running artificial intelligence models (so-called inference) rather than training artificial intelligence. Smart models, the ACDC project has been in the works for a few years, and it's not yet certain when the new chips will be launched, if at all. Apple promised a number of new artificial intelligence products and announcements at its Worldwide Developers Conference in June.
Learn more:
03. The largest artificial intelligence financing in British history: Wayve raises $1.05 billion to bring its Tesla-like self-driving technology to many car manufacturers
Wayve, a British-born startup developing autonomous driving systems, has completed a $1.05 billion Series C round of financing led by SoftBank Group. It is the largest ever AI fundraising event in the UK and one of the top 20 AI fundraising events in the world to date. The company plans to use the new capital infusion to develop its products for fully autonomous driving and other AI-assisted automotive applications. Wayve's approach to self-driving is similar to Tesla's, but Wayve plans to sell its self-driving models to various automotive OEMs. Of course, this means Wayve will get more training data to improve its models.
Learn more:
04. Zhipu AI is developing a domestic Wensheng video model to benchmark Sora, which will be released as soon as this year.
Zhipu AI is one of the representatives of the leading domestic model manufacturers and has built an AIGC model product matrix. Zhipu AI is developing a high-quality Wensheng video model to benchmark Sora. Key breakthroughs include the selection of technical routes and the training of high-quality video material corpus. The goal is to achieve super cognitive intelligence that exceeds human levels while ensuring that the model is safe and controllable. It has developed a domestic Wensheng video model that is benchmarked against Sora and is expected to be released as soon as this year.
Learn more:
https://www.tmtpost.com/7072424.html
05.Microsoft uses MAI-1 to develop the latest model and will compete with Google, Anthropic and OpenAI for the most advanced models
With MAI-1, Microsoft is developing a new internal AI model with about 500 billion parameters that will compete with state-of-the-art models from Google, Anthropic, and OpenAI. Microsoft is pursuing a dual strategy of developing new models for mobile applications. Develop "small language models" as well as build for larger, more powerful AI models. The exact use of MAI-1 depends on its performance. A preview version may be shown at the Build developer conference at the end of May.
Learn more: https://the-decoder.com/microsoft-allegedly-developing-mai-1-a-competing-model-to-openais-gpt-4/
Article recommendations
From attention mechanism to RLHF, a must-read list for getting started with large model technology
7 prompt tips to make your conversation with AI more effective
If there is any infringement, please contact us for deletion.
Follow us
OpenSPG:
Official website: https://spg.openkg.cn
Github: https://github.com/OpenSPG/openspg
OpenASCE:
官网：https://openasce.openfinai.org/
GitHub：[https://github.com/Open-All-Scale-Causal-Engine/OpenASCE]