Technology peripherals
AI
The world's first AI safety research institute will be established in the UK
The world's first AI safety research institute will be established in the UK
IT House News on October 27, based on reports from CNBC, Reuters and other reports, on Thursday local time, British Prime Minister Sunak announced plans to establish the world’s first AI security research institute, which will be launched on November 1-2 AI Security Summit will be held today. The summit will bring together AI companies, governments, civil society and experts in related fields from around the world to discuss how to reduce the risks caused by AI through international coordinated action.
Image source Pexels
Sunak said in his speech that
This soon-to-be-established institute will promote the world’s understanding of AI security, and will carefully research, evaluate and test new AI technologies in order to understand each new model capabilities and explore the full range of risks from social harms such as “bias and misinformation” to the “most extreme risks”. Sunak said, “
AI can make it easier to create chemical or biological weapons, but this technology can also be exploited by criminals.In the most unlikely but most extreme circumstances , it is even possible for humans to completely lose control in the face of AI through AI, which is sometimes called super intelligence.” In addition, at the AI Security Summit to be held next week, Sunak will propose the establishment of a "truly global expert group" -
Members of this group will be nominated by participating countries and organizations, and will also be A report on the state of AI science will be released. According to previous reports from IT House, the British antitrust regulator CMA (Competition and Markets Authority) last month proposed new standards for managing AI models, covering matters such as accountability, access and transparency. Among the regulatory principles announced by the CMA at that time,
included that AI developers and deployers should be responsible for output content and must not engage in anti-competitive behavior.
The above is the detailed content of The world's first AI safety research institute will be established in the UK. For more information, please follow other related articles on the PHP Chinese website!
Hot AI Tools
Undresser.AI Undress
AI-powered app for creating realistic nude photos
AI Clothes Remover
Online AI tool for removing clothes from photos.
Undress AI Tool
Undress images for free
Clothoff.io
AI clothes remover
Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!
Hot Article
Hot Tools
Notepad++7.3.1
Easy-to-use and free code editor
SublimeText3 Chinese version
Chinese version, very easy to use
Zend Studio 13.0.1
Powerful PHP integrated development environment
Dreamweaver CS6
Visual web development tools
SublimeText3 Mac version
God-level code editing software (SublimeText3)
Hot Topics
The terrifying robot Ameca can talk, all thanks to GPT-3!
Apr 12, 2023 am 09:46 AM
In addition to making "emoticons", Ameca, the world's most advanced humanoid robot, can speak. Recently, the British robotics company Engineered Arts released a latest video showing a conversation between Ameca and some engineers. When faced with questions from engineers, its answers were clear and logical. The most important thing is that the expression, eyes and movements when it answers give people a sense of intimacy in communication. AI has become conscious, and the doomsday scenes of "Terminator" and "I, Robot" are about to come. You don't have to worry about this. Ameca said, "Robots will never rule the world. We are here to help and serve humans, not replace them." Poetry is not poetry, music is not music. This robot can speak, all based on speech synthesis.
Samsung acquires British knowledge graph startup Local AI simulates human thinking to handle tasks
Jul 19, 2024 pm 12:44 PM
Recently, Samsung announced the acquisition of Oxford Semantic Technologies, a British knowledge graph startup, to enhance its local AI capabilities and provide users with a more personalized AI experience. The company's main product is the AI engine RDFox, which uses knowledge graph technology to store information as an interconnected network. The way it processes data is similar to how humans think: acquiring, memorizing, recalling and reasoning about knowledge. This technology will enhance the device's understanding of the products or services used by users, thereby enabling rapid information retrieval and recommendations. It is understood that Oxford Semantic Technologies was founded in 2017 by three Oxford University professors: Ian Horrocks, Boris Mortick and Bernardo Cuenca.
In the AI era, how to use ChatGPT safely has triggered heated discussions
Jun 03, 2023 pm 05:35 PM
Since its public launch in November 2022, ChatGPT has experienced significant growth. It has become an indispensable tool for many businesses and individuals, but as ChatGPT is integrated into our daily lives and work on a large scale, people will naturally think: Is ChatGPT safe to use? ChatGPT is generally considered safe due to the extensive security measures, data handling methods, and privacy policies implemented by its developers. However, like any technology, ChatGPT is not immune to security issues and vulnerabilities. This article will help you better understand the security of ChatGPT and AI language models. We will explore aspects such as data confidentiality, user privacy, potential risks, AI regulation and security measures. Finally, you will be interested in Chat
The United States and the United Kingdom sign an agreement to jointly develop artificial intelligence safety standards
Apr 04, 2024 pm 02:31 PM
The two countries plan to conduct at least one joint test of artificial intelligence models. On April 1, 2024, the United States and the United Kingdom reached a memorandum of understanding (MOU) to jointly develop artificial intelligence (AI) safety standards. A core goal of this agreement is to develop and continuously iterate a "robust evaluation suite" for AI models, systems, and agents through a coordinated scientific approach. Specifically, the two countries plan to conduct at least one joint test of a publicly accessible AI model. In addition, the two parties also plan to conduct personnel exchanges between the previously established AI Security Research Institute. The partnership is effective immediately. The U.S. Department of Commerce and the U.K. Department of Science, Innovation and Technology announced the partnership in the same press release. Safe AI development under review of this new
Is it possible to gain a sense of technology by paying homage to the Queen's remains? Using a robotic arm to hold the coffin and rotate it 360 degrees, it only takes 3 days to serve 67 million people
Apr 08, 2023 pm 05:31 PM
Produced by Big Data Digest Author: Caleb On September 19, Queen Elizabeth II of the United Kingdom and the Commonwealth of Nations held a ceremony at Westminster Abbey in London. At 4 p.m. local time, the queen's burial ceremony was held at St. George's Church. The coffin will be temporarily placed in the royal tomb until the queen's family members hold a family memorial ceremony in the local evening. Finally, the Queen will be buried together with her husband, Prince Philip, who passed away last year, in the George VI Memorial Chapel in St. George's Church. Before the state funeral, the Queen's body was moved from Buckingham Palace to Westminster Abbey for people in the UK and around the world to pay their respects. Last Friday, Beckham also queued for 12 hours just to say his final farewell to the Queen. However, because there were so many people coming to express their condolences, officials even suspended the queue from time to time, and the entire waiting process was
UK makes big move into cryptocurrency: first exchange-traded product is about to launch
Jun 05, 2024 pm 06:11 PM
The United Kingdom’s Financial Conduct Authority (FCA) has officially approved the first batch of cryptocurrency exchange-traded products (ETPs), marking an important development in integrating digital assets into the country’s financial landscape. The move brings the UK into line with global financial centers that have embraced trading in this suite of digital assets. Regulatory Background and International Comparisons WisdomTree has announced that it has received approval from the FCA to list two physically-backed crypto ETPs tracking Bitcoin and Ethereum on the London Stock Exchange. The products are scheduled to begin trading on May 28, according to Bloomberg, underscoring the rapid progress from regulatory approval to market availability. WisdomTree's move is part of a wider trend that has seen other issuers such as ETC Group, 21S
When AI encounters fraud
May 31, 2023 pm 02:06 PM
"Although I know that scams are common nowadays, I still can't believe that I have actually encountered one." On May 23, reader Wu Jia (pseudonym) was still frightened when recalling the telecom fraud she encountered a few days ago. In the scam Wu Jia encountered, the scammer used AI to change her face into someone she was familiar with. Not only did Wu Jia encounter AI fraud that was difficult to guard against in her daily life, a reporter from Beijing Business Daily noticed that new telecommunications fraud models using AI technology have shown a high incidence in recent times. “The success rate of AI fraud is close to 100%.” “Technology Topics such as "Company owner was defrauded of 4.3 million yuan in 10 minutes" have been on the hot search topics one after another, which has also triggered discussions among users on the application of new technologies. "AI face-changing" fraud Artificial intelligence has become popular again, this time around telecommunications fraud. Wu Jia
A long article of 10,000 words丨Deconstructing the AI security industry chain, solutions and entrepreneurial opportunities
Jun 06, 2023 pm 12:53 PM
Key points: 1. The security issue of large AI models is never a single issue. Just like human health management, it is a complex and systematic system engineering involving multiple subjects and the entire industry chain. 2. AI security is divided into: the security of large language models (AISafety), the security of models and usage models (Security for AI), and the impact of the development of large language models on existing network security, which corresponds to individual security, environmental security and social security. different levels. 3. AI, as a "new species", requires safety monitoring during the training process of large models. When the large models are finally introduced to the market, they also need a "quality inspection". After quality inspection, they enter the market and need to be used in a controlled manner. methods, these are all macros to solve security problems
