AstraZeneca in deal with China firm worth up to £4bn
Britain's largest listed company announced the tie-up with biotech firm CSPC Pharmaceuticals following controversy over its involvement in China.
The pharmaceuticals giant will pay CSPC an upfront fee of £81m while a total further payment of up to £3.8billion is available if the drugs reach development and sales-related milestones.
AstraZeneca has been ploughing cash into the country, including announcing a £1.8billion research and development hub in the capital Beijing.
The firm's former boss in China, Leon Wang, was arrested in October and is still thought to be in detention.
More recently, former Tory leader Sir Iain Duncan Smith has accused the pharma giant of 'turning a blind eye to the nature of the Chinese government'.
AstraZeneca boss Sir Pascal Soriot said in February: 'We all think about Leon and miss him, but the reality is we are not able to talk to him. We are not allowed.'
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Reuters
24 minutes ago
- Reuters
China's Baidu misses quarterly revenue estimates on advertising weakness
Aug 20 (Reuters) - Chinese search engine giant Baidu's ( opens new tab, quarterly revenue fell and was just shy of market estimates on Wednesday, signaling persistent weakness in the advertising market amid prolonged economic uncertainty. U.S.-listed shares of the company fell about 2% in premarket trading. Slammed by a property market downturn, weak employment rates and choppy consumer demand, companies in the world's second-largest economy have reined in advertising spending to cut costs and protect their margins. Baidu, which relies heavily on advertising in its search engine, also took a hit. Its core online advertising business, which typically makes up 60% of the overall company revenue, reported a 15% decrease in revenue to 16.2 billion yuan during the April-June quarter. That eclipsed upbeat growth at the company's AI cloud segment, which drove a 34% increase in Baidu's non-online advertising business. The company reported total revenue of 32.71 billion yuan ($4.56 billion) during the second quarter, down 4% from the year earlier. Analysts on average estimated 32.76 billion yuan, according to data compiled by LSEG. ($1 = 7.1784 Chinese yuan renminbi)
Geeky Gadgets
an hour ago
- Geeky Gadgets
GPT-5 Pro : The $200 AI Tool That's Smarter Than You Think
What if the future of innovation wasn't just faster or smarter, but fundamentally redefined? Enter GPT-5 Pro, the latest leap in artificial intelligence that doesn't just assist, it thinks, reasons, and solves like never before. Imagine a tool capable of automating complex coding, crafting hyper-intelligent AI agents, and integrating seamlessly into your workflows, all while boasting an IQ comparable to Mensa-level intelligence. It's not just a technological upgrade; it's a paradigm shift. Yet, with its $200 monthly price tag, this new tool raises a provocative question: Is the future of AI reserved for the few who can afford it, or will it become a shared resource for all? David Ondrej explore how GPT-5 Pro is reshaping industries and empowering professionals to achieve what was once thought impossible. From automating tedious tasks to unlocking new entrepreneurial opportunities, this AI powerhouse offers tools that could transform how we work and innovate. But it's not without its challenges, its high cost and potential to widen the digital divide demand attention. Whether you're a developer eyeing its coding prowess or a business leader seeking to streamline operations, GPT-5 Pro offers a glimpse into the future of AI-driven transformation. The question is: How will you harness it? GPT-5 Pro Overview Key Features and Distinctive Capabilities GPT-5 Pro stands out due to its ability to process multiple reasoning chains simultaneously, allowing it to solve complex problems with remarkable efficiency. With an IQ score comparable to Mensa-level intelligence (148), the model excels in logic-based tasks, coding automation, and decision-making. These features make it a powerful resource for professionals seeking to streamline workflows or develop innovative solutions. Some of its standout features include: Automating coding tasks: GPT-5 Pro can write, debug, and optimize code, significantly reducing development time. GPT-5 Pro can write, debug, and optimize code, significantly reducing development time. Creating AI voice agents: Using Retrieval-Augmented Generation (RAG), it enables the development of highly responsive and context-aware AI agents. Using Retrieval-Augmented Generation (RAG), it enables the development of highly responsive and context-aware AI agents. Seamless integration: The model integrates effortlessly with tools like OpenAI's real-time API and GitHub, enhancing application development efficiency. For example, developers can use GPT-5 Pro to build sophisticated applications with minimal effort, while businesses can automate repetitive processes, saving both time and resources. Its adaptability ensures that it meets the diverse needs of professionals across various industries. Applications Across Industries The versatility of GPT-5 Pro makes it a fantastic tool across multiple sectors, particularly in business automation, project development, and entrepreneurial ventures. Entrepreneurs can use its capabilities to launch AI-driven businesses, automate routine tasks, and optimize operational workflows. Freelancers, on the other hand, can tap into high-value opportunities, such as creating AI-powered solutions that command premium fees on platforms like Upwork. Additionally, GPT-5 Pro's iterative problem-solving approach allows users to refine their projects over time, addressing challenges as they emerge. This feature is particularly beneficial for scaling businesses, allowing them to generate substantial revenue while minimizing operational complexities. Its ability to adapt and improve ensures that it remains a valuable asset for long-term growth and innovation. GPT-5 Pro Review : Is a $200 AI Tool Worth the Investment? Watch this video on YouTube. Learn more about OpenAI's GPT-5 with the help of our in-depth articles and helpful guides. Challenges of Accessibility Despite its advanced capabilities, GPT-5 Pro's high subscription cost, currently set at $200 per month, poses a significant barrier to entry for many users. This pricing model risks creating a 'pay-to-play' environment, where only those with substantial financial resources can access the most sophisticated AI tools. Such disparities could exacerbate existing inequalities, limiting opportunities for smaller businesses and independent developers to compete in an increasingly AI-driven economy. Addressing these accessibility challenges is essential to making sure that the benefits of AI are distributed more equitably. Without deliberate efforts to make advanced AI tools like GPT-5 Pro more affordable and accessible, the technology's potential to drive widespread innovation may remain unrealized. Strategies for Maximizing GPT-5 Pro's Potential To fully harness the capabilities of GPT-5 Pro, a structured and strategic approach to implementation is crucial. Consider the following strategies to optimize its use: Master prompt engineering: Craft precise and well-structured prompts to communicate effectively with the model and achieve accurate, relevant results. Craft precise and well-structured prompts to communicate effectively with the model and achieve accurate, relevant results. Use integration tools: Use platforms like OpenAI's API and GitHub to seamlessly incorporate GPT-5 Pro into your workflows and applications. Use platforms like OpenAI's API and GitHub to seamlessly incorporate GPT-5 Pro into your workflows and applications. Enhance outputs with RAG: Use Retrieval-Augmented Generation to improve the relevance and accuracy of the model's responses, particularly for specialized tasks. Use Retrieval-Augmented Generation to improve the relevance and accuracy of the model's responses, particularly for specialized tasks. Adopt iterative problem-solving: Continuously refine your projects by addressing challenges and incorporating feedback, making sure long-term success and adaptability. By employing these techniques, you can unlock the full potential of GPT-5 Pro, allowing you to achieve your objectives efficiently while staying ahead in a competitive landscape. Future Implications and Considerations As artificial intelligence continues to evolve, tools like GPT-5 Pro are poised to play an increasingly central role in shaping industries and driving innovation. Early adoption of such technologies can provide a significant competitive advantage, allowing businesses and professionals to stay ahead in a rapidly changing market. Developing strong problem-solving skills and integrating AI into core business strategies will be essential for long-term success. However, the growing reliance on advanced AI also highlights the need for a more inclusive ecosystem. Efforts to address accessibility challenges will be critical to making sure that the benefits of AI are shared broadly, fostering innovation across diverse sectors and communities. Policymakers, developers, and organizations must collaborate to create frameworks that promote equitable access to AI technologies, allowing a more dynamic and inclusive future. GPT-5 Pro represents a remarkable step forward in artificial intelligence, offering capabilities that have the potential to transform industries and redefine professional workflows. By adopting effective implementation strategies and advocating for greater accessibility, you can use this powerful tool to drive innovation, achieve your goals, and contribute to a more equitable and technologically advanced world. Media Credit: David Ondrej Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
Geeky Gadgets
an hour ago
- Geeky Gadgets
Meet Deepagents : The AI That Builds Its Own Problem-Solving Team
Imagine an AI system so adaptable it can break down intricate problems into manageable pieces, delegate tasks to specialized sub-agents, and dynamically adjust its approach based on real-time data. This isn't science fiction, it's the promise of deepagents. Built on the innovative Langraph framework, these agents represent a leap forward in artificial intelligence, combining advanced planning with modular tools to tackle challenges that traditional systems struggle to solve. Whether you're managing massive datasets, automating complex workflows, or exploring new frontiers in research, deepagents offer a level of precision and scalability that feels almost futuristic. But how do you go from concept to implementation? That's where this quick-start guide comes in. Below LangChain explain how to implement deepagents step by step, from installation to customization. You'll learn how to harness the power of the react agent loop, a mechanism that enables iterative decision-making, and explore tools like the sub-agent system, which allows for seamless task delegation. Along the way, we'll highlight practical tips for tailoring deepagents to your unique needs, whether that means creating custom tools or optimizing state management for your workflows. By the end, you won't just understand deepagents, you'll be ready to build and deploy them to solve real-world problems. Let's explore what's possible when intelligence meets adaptability. Deepagents Overview Understanding Deepagents Deepagents are engineered to solve intricate problems that demand extended planning and adaptive problem-solving. Operating on the Langraph framework, they create agent graphs that streamline communication and task delegation. At the heart of their operation is the 'react agent' loop, a mechanism that enables agents to make iterative decisions based on real-time data and evolving contexts. By integrating modular tools and sub-agents, deepagents can break down complex tasks into manageable components, making sure efficiency and precision. This adaptability makes them suitable for a wide range of applications, from research and content generation to data analysis and beyond. State Management in Deepagents Effective state management is a fundamental aspect of deepagents, making sure that all interactions, tasks, and data are tracked seamlessly. The agent's state is composed of three primary elements: Messages: Tracks human inputs, AI responses, and outputs from integrated tools, making sure a clear record of communication. Tracks human inputs, AI responses, and outputs from integrated tools, making sure a clear record of communication. To-Dos: Manages tasks with statuses such as pending, in-progress, and completed, providing a structured workflow. Manages tasks with statuses such as pending, in-progress, and completed, providing a structured workflow. Files: Operates within a virtual file system, represented as dictionaries. This system supports scalability and parallel processing, allowing the agent to handle large-scale data efficiently. The virtual file system is particularly noteworthy, as it allows deepagents to manage complex workflows and large datasets with ease. This capability is essential for tasks that require high levels of accuracy and organization. How to Implement Deepagents Watch this video on YouTube. Here are more detailed guides and articles that you may find helpful on AI Agents. Core Tools and Functionalities Deepagents are equipped with a robust suite of tools that enhance their functionality and adaptability. These tools include: Planning Tool: Organizes tasks into distinct states, allowing clear progress tracking and efficient task management. Organizes tasks into distinct states, allowing clear progress tracking and efficient task management. File System Tools: Provides operations such as reading, writing, listing, and editing files. Advanced features include line offsets, truncation, and a file reducer for merging changes through parallel processing. Provides operations such as reading, writing, listing, and editing files. Advanced features include line offsets, truncation, and a file reducer for merging changes through parallel processing. Sub-Agent Tool: Assists the creation of independent sub-agents with isolated states. These sub-agents focus on specific tasks and return only final results to the main agent, making sure streamlined operations. These tools work in unison to empower deepagents to tackle diverse and complex tasks with precision. The modular nature of these tools allows for customization, making it possible to adapt the agent's capabilities to specific requirements. Tailoring Deepagents to Your Needs One of the most compelling aspects of deepagents is their high degree of customization. Users can define custom tools, instructions, models, and sub-agents to tailor the agent's functionality to specific use cases. While default tools and models, such as Claude, are provided, they can be replaced or extended to meet unique requirements. Additionally, state schemas can be modified to track attributes relevant to specific tasks. This flexibility ensures that deepagents can be adapted to a wide range of applications, from simple workflows to highly specialized projects. Steps to Implement Deepagents Implementing deepagents involves a straightforward process. Follow these steps to get started: Install the Package: Use pip install deepagents to install the necessary components. Use to install the necessary components. Define Custom Tools and Instructions: Create tools and instructions tailored to your specific tasks and objectives. Create tools and instructions tailored to your specific tasks and objectives. Create a Deepagent: Use the create_react_agent function to combine built-in and custom components into a cohesive agent. Use the function to combine built-in and custom components into a cohesive agent. Invoke the Agent: Deploy the agent for tasks such as research, content generation, or data analysis. Sub-agents can also be defined with specific tools and instructions, allowing them to operate independently while contributing to the overall task. This modular approach ensures that each component of the agent is optimized for its specific role. Addressing Challenges and Enhancing Capabilities While deepagents are highly capable, there are areas where further refinement is necessary to enhance their functionality. Key challenges include: File Merging: Handling edge cases, such as simultaneous edits to the same file, requires more robust solutions to ensure data integrity. Handling edge cases, such as simultaneous edits to the same file, requires more robust solutions to ensure data integrity. Expanded Functionality: Developing additional tools and features will enable deepagents to address increasingly complex use cases, improving their scalability and versatility. Ongoing advancements in the Langraph framework and related technologies will play a crucial role in overcoming these challenges. By addressing these areas, deepagents can continue to evolve and remain at the forefront of AI development. Using Deepagents for Complex Tasks Deepagents provide a powerful, modular framework for building intelligent agents capable of tackling intricate tasks with advanced planning and tool integration. By using the Langraph framework, customizable tools, and sub-agent functionality, users can create scalable AI solutions tailored to their specific needs. Whether applied to research, content generation, or data processing, deepagents offer a robust platform for addressing complex challenges in AI development. Media Credit: LangChain Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
