Anthropic has signed a $10 billion deal with AI cloud startup Volta to expand its computing capacity. The deal will help Anthropic keep pace with demand for its products, including its chatbot Claude. Volta will develop a data center in Norway with a 133 megawatt capacity, powered by crypto-mining company Bitdeer.
Nvidia, a leader in AI model training, is poised to benefit from the AI infrastructure spending spree, along with companies like SK Hynix and TSMC. LinkedIn, on the other hand, has decided to focus on improving efficiency and getting more out of its existing hardware rather than expanding its AI data centers.
In the US Congress, ChatGPT has become the dominant AI tool, accounting for nearly $9 out of every $10 spent on identifiable AI tools. Anthropic's Claude is a distant second in terms of spending. Meanwhile, researchers have introduced FLEX, a new benchmark for multimodal continual instruction tuning, aimed at improving the performance of large language models.
Cybersecurity threats are also on the rise, with cybercriminals using trojanized GitHub repositories to distribute malware to AI developers. A report from INTERPOL has found that AI is driving over 55% of cybercrime in Africa, highlighting the growing threat of AI-powered cybercrime.
In the UK, the regulator is considering whether AI scribes should be classified as medical devices, which could have implications for the use of AI in healthcare. Some US AI leaders are turning to Chinese open-weight models, citing concerns about safety and security.
Key Takeaways
['Anthropic signs $10 billion deal with AI cloud startup Volta to expand computing capacity.', 'Nvidia, SK Hynix, and TSMC poised to benefit from AI infrastructure spending surge.', 'ChatGPT dominates AI spending in US Congress, accounting for nearly 90% of spending.', "Anthropic's Claude is a distant second in AI spending in US Congress.", 'Researchers introduce FLEX, a new benchmark for multimodal continual instruction tuning.', 'LinkedIn focuses on AI efficiency rather than hardware expansion.', 'UK regulator considers classifying AI scribes as medical devices.', 'INTERPOL report finds AI linked to over 55% of cybercrime in Africa.', 'US AI leaders turn to Chinese open-weight models citing safety and security concerns.', 'Cybercriminals use trojanized GitHub repositories to distribute malware to AI developers.']Anthropic signs $10 billion deal for computing capacity
Anthropic has struck a $10 billion deal with a new cloud startup for computing capacity. The startup, which has only been in operation for a few months, will help Anthropic keep pace with demand for its products. The deal marks Anthropic's latest effort to expand its computing capacity. The startup will develop a data center in Norway with a 133 megawatt capacity.
Anthropic signs $10 billion deal with AI cloud startup Volta
Anthropic has signed a $10 billion deal with AI cloud startup Volta. The deal will help Anthropic expand its computing capacity. Volta will develop a data center in Norway with a 133 megawatt capacity. The data center will be powered by a partner, Bitdeer, a crypto-mining company.
New study on multimodal continual instruction tuning
Researchers have introduced FLEX, a new benchmark for multimodal continual instruction tuning. The study aims to improve the performance of large language models in acquiring new tasks while retaining previously learned capabilities. The researchers found that current methods for expert routing are nearly saturated, making it challenging to improve performance.
ChatGPT dominates AI spending in US Congress
ChatGPT has become the dominant AI tool in the US Congress, accounting for nearly $9 out of every $10 spent on identifiable AI tools. The chatbot has been used for tasks such as summarizing legislation, writing memos, and preparing hearing materials. Anthropic's Claude is a distant second in terms of spending.
3 stocks to benefit from AI infrastructure surge
The AI infrastructure spending spree is expected to continue, with companies like Nvidia, SK Hynix, and TSMC poised to benefit. Nvidia is a leader in AI model training, while SK Hynix is a major supplier of high-bandwidth memory. TSMC is working to build capacity to meet surging demand.
LinkedIn to focus on AI efficiency rather than hardware
LinkedIn has decided not to spend aggressively on expanding its AI data centers. Instead, the company will focus on improving efficiency and getting more out of its existing hardware. LinkedIn's CTO says the company has found ways to double the performance of its existing GPUs.
UK regulator weighs in on AI scribes as medical devices
The UK regulator is considering whether AI scribes should be classified as medical devices. The decision could have implications for the use of AI in healthcare. AI scribes are being used to help with medical transcription and other tasks.
AI developers targeted by trojanized GitHub repositories
Cybercriminals are using trojanized GitHub repositories to distribute malware to AI developers. The malware is designed to steal sensitive information and can be used to carry out further attacks. The attackers are using fake AI tools and repositories to trick developers into downloading the malware.
INTERPOL report finds AI linked to over half of cybercrime in Africa
A new report from INTERPOL has found that AI is driving over 55% of cybercrime in Africa. The report highlights the growing threat of AI-powered cybercrime and the need for international cooperation to combat it.
US AI leaders turn to Chinese open models
Some US AI leaders are turning to Chinese open-weight models, citing concerns about safety and security. The move challenges the claims of closed-source AI model developers that open-source models pose a threat to society.
Sources
- Anthropic Inks $10 Billion Computing Deal With New Cloud Startup
- Anthropic signs $10 billion deal with AI cloud startup Volta
- Beyond Routing Saturation: A Long-Horizon Class-Incremental Perspective on Expert Routing in Multimodal Continual Instruction Tuning
- What Is Congress’s Favorite AI Tool?
- 3 Stocks to Buy as AI Infrastructure Continues to Surge
- LinkedIn says it won't be spending big on AI hardware this year — but it has a good reason why
- Are AI scribes medical devices? UK regulator weighs in.
- AI developers targeted via trojanized GitHub repositories
- INTERPOL report finds AI linked to over half of cybercrime in Africa
- US AI leaders turn to Chinese open models, challenging closed AI safety claims
Comments
Please log in to post a comment.