HawkInsight

  • Contact us
  • App
  • English

2023 World Artificial Intelligence Conference Opens in Shanghai Musk "Surprise Appearance"

On the morning of July 6, the 2023 World Artificial Intelligence Conference (WAIC) opened at the Shanghai World Expo Center and Expo Exhibition Hall。The theme of this conference is "Zhilian World Generates Future," and the number of exhibitors and exhibition area are the highest ever.。

On the morning of July 6, the 2023 World Artificial Intelligence Conference (WAIC) opened at the Shanghai World Expo Center and Expo Exhibition Hall。

According to the official website, the conference will be held for three days, the number of exhibitors and exhibition area are the largest ever。The theme of this conference is "Zhilian World Generates the Future," and the overall structure of the conference activities is set up according to "1 + 1 + 2 + 10 + N," namely, 1 opening ceremony, 1 closing ceremony, 2 plenary meetings of scientific frontier and industrial development, 10 theme forums, and N ecological forums and activities.。

At the opening ceremony, Tesla founder and CEO Elon Musk "appeared" in a video surprise and delivered a speech。

Musk said that the emergence of generative artificial intelligence is having a very profound effect and impact on human civilization.。With the explosive growth of computer computing power, it means that the gap between the computing power of machines and organisms is further widening, and after a period of time, the proportion of artificial intelligence will become lower and lower, which will be a fundamental change in depth.。

He also mentioned that his humanoid robot is still in the development stage, but there will be many robots in the future.。With the rapid development of artificial intelligence technology, it is expected that FSD will be fully autonomous by the end of this year.。"I've made many similar predictions before, and I admit that the previous predictions were not entirely accurate, but this one, I think, is closer.。"

He also called for regulation of AI.。"We need some regulatory oversight to ensure this kind of deep AI development."。"Such superintelligence has the ability to be more powerful than humans."。It's a dedication and a concern。It may lead to a positive future, but there is also a certain probability of a negative future.。We want to do our best to make sure that these negative futures don't happen, but that a positive future is achieved.。"

马斯克

In the "plenary" stage, there are a number of "big" will appear。Including Tencent Group Senior Executive Vice President and CEO of Cloud and Smart Industry Group Tang Daosheng, Ant Group Vice President and General Manager of Digital Technology Solutions Yu Hu, AMD Chairman and CEO Su Zifeng, Qualcomm Technologies Chief Commercial Officer Jim Casey and Jingdong Group Vice President and Dean of Jingdong Exploration Research Institute He Xiaodong and many others will attend。

It is reported that more than 1,000 domestic and foreign scholars, well-known entrepreneurs and representatives of international organizations will attend the exhibition this year.。So, with so many people gathered, what are the highlights of this conference??

  

Immersive AI interactive experience

  

The conference has a number of AI interactive experience projects, you can immerse, close to the very leading artificial intelligence technology products and capabilities。Such as NetEase Fuxi self-developed picture generation model - "Danqing," relying on this model can let participants experience AI painting: the audience only need to enter the key words of the work, choose different painting styles, you can generate exquisite paintings, on-site support creative results print souvenirs。

There is also the world's first Text-to-3D product "ChatAvatar," which only needs a text description or a photo to generate video-grade ultra-realistic 3D digital human assets, which can provide audio, video-driven solutions and other digital human technology services, adapted to unity, ue, maya and other platforms, to meet the needs of users in various situations, and further accelerate the artist's creative process.。

WAIC 2023 in conjunction with the West Bank Art Museum to hold the "Temple of Plasma" AI performance, which will be presented through a combination of cultural and technological components.。Among them, the cultural part is expressed through national musical instruments and actors' costumes, while the scientific and technological part is presented through 3D printing technology to complete the production of musical instrument bones and TD software to generate real-time interactive changes.。

In addition to AI interaction, the conference also set up a robot experience area in Minhang area。Once in, there will be robotic receptionists and volunteers。Traditional Chinese medicine robots can provide participants with a variety of diagnostic methods。Participants can also enjoy soccer games and dance performances presented by robots, and robot attendants provide tea delivery and delivery services, as well as security robots to provide the venue to maintain order throughout the process.。

  

A number of large models were unveiled.

  

This year's World Artificial Intelligence Conference set up a special exhibition area on the theme of "Towards General Artificial Intelligence," with "Building a World-Class Industrial Cluster of Artificial Intelligence" as a guide, setting up four major sections: "Computing Power Foundation," "Model Base," "Intelligent Application" and "Innovation Frontier," with a total of 49 enterprises or institutions stationed in the exhibition area.。

● Calculation force foundation plate

With the rapid development of generative AI, the computing power chip market is also in short supply.。The conference will focus on core computing technologies such as smart chips, hardware and software adaptation and training frameworks, and will bring together 12 chip companies in the core exhibition area, bringing a total of 13 chip exhibits.。Including Muxi's "Xisi N100 Artificial Intelligence Reasoning GPU," Hanbo Semiconductor's "Hanbo Second Generation Full Function GPU Chip," Yuntian Li Fei's DeepEdge10 Series Large Model Reasoning Edge Computing Chip, and Flint Technology's "Artificial Intelligence Cloud Training Chip 2.0 "and" Artificial Intelligence Cloud Inference Chip 2.5, "Qualcomm's" Second Generation Xiaolong 8 Mobile Platform "and so on。

● Model base plate

The World Artificial Intelligence Conference will bring together 14 companies to bring the latest results in the field of large models in the basic model section of the core exhibition area.。Including Shang Tang Technology's "Daily New SenseNova Model," Baidu's "Wen Xin Yi Yan," HKUST Xunfei's "Xunfei Spark Cognition Model," 360 Group's "360 Zhinao," Fudan University's Natural Language Processing Laboratory's "MOSS," Huawei's "Huawei Cloud Disk Ancient Model," Alibaba's "Tongyi Thousand Questions," Jingdong Cloud's "Sharp Speech Model," etc.。

● Intelligent application plate

11 companies will showcase their latest achievements in the vertical field of large models in the core exhibition area of this world artificial intelligence conference。For example, Jinshan Office brings artificial intelligence office application "WPS AI" with large language model capability; Star Ring Technology's "Star Ring Endless Infinity Financial Model"; Verizon's new "CityGPT," Shanghai Artificial Intelligence Laboratory's "General Model System for Scholars" and so on.。In addition to Chinese companies, other well-known foreign companies have also emerged with large models, such as Amazon Web Services's "Amazon Bedrock" and Qualcomm's "End-Side Generative AI Use Case Demonstration."。

● Innovation frontier plate

In this exhibition area, there will be a large number of innovative large model applications in the frontier field.。

In the AI4S field, Jingtai Zhiyao will show the demo model of "Jingtai Technology Intelligent Automation Laboratory" for the first time.。This is its self-developed automated digital intelligence laboratory, which has been successfully applied in pharmaceutical, chemical, testing, functional materials, bio-based materials and other industries.。As the leader and practitioner of the "AI for Science" research paradigm, Beijing Deep Technology will launch a new generation of micro-scale industrial design and simulation platform at this conference, and the company will be committed to creating a new paradigm of "computation-guided experiments and experimental optimization design."。

In the field of brain-like intelligence, the Institute of Brain-like Intelligence Science and Technology of Fudan University will bring the world's first digital twin brain of 86 billion neurons.。Fudan University research team research and development started, using high-precision human brain multi-modal neuroimaging data, on 14,012 GPU (57.39 million cores) supercomputer cluster, using data assimilation to reconstruct the pulse neural network method, to build a digital twin brain with the size of 86 billion neurons.。

In the field of physical intelligence, the "humanoid robot" of Shanghai Institute of wisdom, Tesla's "Tesla humanoid robot," Fu Liye "universal humanoid robot GR-1," Tongji University's "humanoid robot multi-mode interaction," Shanghai University of Technology "Beckham robot 4.0, "Netease Fuxi" Body Intelligence-Unmanned Loader "will also meet with the public at the conference。

世界人工智能大会

  

WAIC 2023 Named Nine "Treasures of Town Hall"

  

The conference selected innovative exhibits with outstanding contributions and leading technologies from the exhibitors, and finally selected nine "treasures of the town hall."。

● Ant Science and Technology Group's Ant Jian AI Safety Detection Platform 2.0

Ant Jian AI Security Detection Platform 2.0 has a one-stop model detection platform, data-based security evaluation system and industrial-level scenario detection solutions, has been applied to finance, insurance, security, government, intelligent customer service and other 10 + scenarios.。

● Amazon Bedrock at Amazon

Amazon Bedrock is a fully managed generative AI service launched by Amazon Web Services, providing basic models from leading AI startups and Amazon self-developed through APIs to help customers easily and securely build and scale generative AI applications.。Customers can choose from a variety of basic models to find the model that best suits their business scenarios.。

● Baidu's Wen Xin Yi Ge

The first domestic technology self-research, based on the large model of Wenxin general and controllable Wensheng graph system.。Wenxinyige can not only assist the artistic creation of visual content creators such as painters and designers, but also provide high-quality and efficient mapping for the media, authors and other text content creators.。

● Proposed untechnical Graphcore ® C600 IPU processor PCIe card

Graphcore C600 IPU processor PCIe card focuses on reasoning and training, which can support various mainstream AI applications and has unique advantages in search and recommendation services.。While providing powerful computing power, ease of use and flexibility, the C600 also achieves low latency and low power consumption. The thermal design power consumption is only 185 watts when running typical workloads, which can greatly reduce data center operating expenses for operation and maintenance personnel。

● Huawei's Teng AI "Big Model Super Factory"

Teng AI "Big Model Super Factory" covers the whole process of big model data & model preparation, calculation power preparation & model training, model deployment online & integration, etc.。

At present, more than 20 domestic large models including Pengcheng Pangu, the industry's first Chinese NLP large model with 200 billion parameters, Zidong Taichu, the industry's first multi-modal large model, and Huawei Yunpangu series have been incubated based on Xunteng AI.。At the same time, Teng AI has also been adapted to support dozens of mainstream open source models in the industry, such as ChatGLM, LLaMA, GPT-3, BLOOM, etc.。

● Jingtai Technology's AI drug development automation solution

Jingtai's AI drug development automation solution realizes a high degree of automation and intelligence in the AI drug development experiment process: it not only breaks through the key speed-limiting bottleneck of chemical synthesis in the DMTA cycle of drug development, but also improves the efficiency and success rate of drug development experiments, and further solves the problem of missing standardized process data in AI drug development experiments.。

● Cloud Sui Intelligent Computing Cluster of Shanghai Fanyuan Technology

The cloud-based intelligent computing cluster is a high-performance artificial intelligence acceleration cluster for large-scale intelligent computing centers, designed to meet the demand for hyperscale computing power in the AIGC era.。At present, based on this cluster product, it has been trained on a kilocalorie scale in the national key laboratory Zhijiang, providing more than 100P of advanced AI computing power to efficiently support multiple AIGC applications including media fusion, text generation PPT applications, cross-modal image generation, etc.。Yunsui intelligent computing cluster can be widely used in smart city computing power base, research institutes to develop computing power clusters, and to meet the standardized AI computing power needs of other enterprises and institutions.。

● Shanghai Xijing Technology's intelligent power exchange unmanned commercial vehicle Q-Truck

Q-Truck is the world's first intelligent electric driverless commercial vehicle, supporting the intelligent energy service PowerOnair, 5 minutes to complete the unmanned battery recharge, suitable for sea, land, air, factories and other logistics scenarios。Q-Truck also appeared in the "Wandering Earth 2" movie as the only driverless truck.。

● Tencent Multimedia Lab XMusic

Tencent Multimedia Lab's latest XMusic Generative Universal Composition Framework based on AIGC technology supports multi-modal content such as videos, pictures, text, tags, and humming as input cues to generate high-quality music with controllable mood, style, and rhythm.。The applicable scenarios of XMusic include video soundtrack, interactive entertainment, auxiliary creation, music education, etc.。

·Original

Disclaimer: The views in this article are from the original author and do not represent the views or position of Hawk Insight. The content of the article is for reference, communication and learning only, and does not constitute investment advice. If it involves copyright issues, please contact us for deletion.