Microsoft unveils GPT-4o for Azure, new AI apps in fight against Google, Amazon
Microsoft's GPT-4o Launch: Leading the AI Race?
Microsoft introduced new AI features at the Build conference, including OpenAI's GPT-4o on Azure and innovative small language models. Amidst a competitive AI race with giants like Google and Amazon, Microsoft also showcased its Cobalt 100 CPU and enhancements to Microsoft 365 with AI copilots for improved productivity. The company's AI-driven strategy significantly contributes to its cloud service growth, positioning Microsoft as a leader in the AI technology sphere.
How does Microsoft's Phi-3-vision model differ from traditional cloud-based AI models?
Microsoft's Phi-3-vision model is designed to run on devices rather than on cloud servers, which sets it apart from traditional cloud-based AI models6. This means that users can access the model in environments where they might not have internet connectivity, making it more versatile and accessible. Additionally, the Phi-3-vision model can accept both image and text prompts, allowing users to ask questions about charts, graphs, or images without needing to connect to the internet.
What capabilities does the new GPT-4o model have, and what controversy arose around its demonstration?
Capabilities of GPT-4o
The new GPT-4o model introduced by OpenAI and showcased by Microsoft at its Build conference is a multimodal AI capable of understanding and responding to inputs via text, audio, and video456. This advanced capability allows it to handle a variety of data inputs more seamlessly compared to previous models6. Microsoft highlighted that developers could now create applications that utilize GPT-4o to interact through images, voice, and text, enhancing the versatility and utility of AI-driven applications4.
Controversy Surrounding the Demonstration
The demonstration of GPT-4o, however, sparked controversy when actress Scarlett Johansson demanded that the company change the voice of its chatbot. This issue arose after the chatbot, which used her voice, was featured in the demonstration answering questions. The controversy highlights the ethical considerations and personal rights issues related to the use of identifiable voices in AI applications.