開発スタッフがアプリ関連の最新技術やニュースをご紹介します。

Google has recently announced a significant enhancement to its Bard AI chatbot, enabling it to provide specific answers to questions related to YouTube videos. While the initial YouTube Extension for Bard, launched in September, allowed users to find specific videos, the latest update empowers the chatbot to respond to queries about the content within videos.

This advancement means users can now engage in richer conversations with Bard, asking detailed questions such as the number of eggs required in a recipe featured in a particular video. Google acknowledges the users’ desire for deeper interaction with YouTube content and aims to fulfill this through Bard’s expanded capabilities.

This development follows YouTube’s recent experimentation with new generative AI features, introducing an AI conversational tool that answers questions about the platform’s content. This tool employs large language models to generate responses, utilizing information from both YouTube and the broader web. Users can now pose questions about the video they are watching, and the AI-driven conversation unfolds in real-time alongside the video playback. Additionally, YouTube has introduced a comments summarizer tool, leveraging generative AI to organize and summarize topics discussed in video comments, providing users with an overview of community discussions.

Coinciding with these updates, Google has expanded access to Bard for teenagers in most countries globally. In a blog post, Google highlighted the potential for teens to use Bard as a tool for inspiration, discovering new hobbies, and solving everyday problems. Whether seeking advice on university applications or exploring leisure activities, teens can now tap into Bard’s capabilities for a wide range of inquiries. This move aligns with Google’s commitment to making Bard a valuable resource for teens across various aspects of their lives.

Yuuma

Tweet

Samsung has introduced its generative AI model, Samsung Gauss, at the Samsung AI Forum 2023. Comprising three tools—Samsung Gauss Language, Samsung Gauss Code, and Samsung Gauss Image—the model aims to enhance productivity across various applications. Samsung Gauss Language is a large language model similar to ChatGPT, capable of understanding and responding to human language. This tool can assist with tasks such as email writing, document summarization, and language translation. Samsung plans to integrate this language model into its devices like phones and laptops. The availability of the model for interaction in English and Korean remains undisclosed.

The Samsung Gauss Code, designed to work with the code assistant code.i, focuses specifically on development code. It aims to assist developers in writing code swiftly, supporting code description and test case generation through an interactive interface. Samsung Gauss Image, as the name implies, focuses on image generation and editing. It can potentially convert low-resolution images into high-resolution ones. While currently limited to internal use, Samsung plans to release Gauss to the public “in the near future,” with a potential application in the Galaxy S24, based on the generative AI model, expected as early as 2024.

Additionally, Samsung has established an AI Red Team to monitor security and privacy concerns throughout the AI development process, ensuring adherence to ethical principles. The company had previously imposed a temporary ban on generative AI tools, including ChatGPT and Google’s Bard, on its devices following an internal data leak.

Samsung’s commitment to generative AI research collaboration with industry and academia was highlighted by Daehyun Kim, the executive vice president of the Samsung Research Global AI Center, at the AI Forum. The name “Samsung Gauss” pays homage to the mathematician Carl Friedrich Gauss, whose normal distribution theory is considered fundamental to AI and machine learning. You can check out the detail event article from Samsaung here.

Yuuma

Tweet

Generative AI systems, which generate content in various formats, are increasingly prevalent across multiple fields like medicine, news, politics, and even providing companionship in social interactions. Initially, these systems primarily produced information in a single format, such as text or graphics, but there is now a notable trend towards enhancing their adaptability to work with additional formats like audio (including voice and music) and video.

The rising usage of generative AI systems underscores the critical need to evaluate potential risks associated with their deployment. As these technologies become more widespread and integrated into diverse applications, concerns about public safety are mounting. Consequently, assessing the potential risks posed by these systems has become a top priority for AI developers, policymakers, regulators, and civil society.

The increasing utilization of these systems underscores the essentiality of evaluating potential dangers linked to the implementation of generative AI systems. Thus, it is becoming increasingly crucial for AI developers, regulators, and civil society to appraise the potential threats these systems might pose. The development of AI that could propagate misinformation raises ethical questions about its societal impact.

In response to these concerns, DeepMind, Google’s AI research lab, has published a paper proposing a framework for assessing the societal and ethical risks associated with AI systems. DeepMind’s proposal emphasizes the necessity for engagement from various stakeholders, including AI developers, app developers, and the general public, in evaluating and auditing AI systems. The research lab underscores the significance of examining AI systems at the “point of human interaction” and understanding their integration into society.

You can checkout the paper here.

Asahi

Tweet

OpenAI has recently established a new team called “Preparedness” to address and assess potential catastrophic risks associated with AI models. This initiative is led by Aleksander Madry, the director of MIT’s Center for Deployable Machine Learning, who joined OpenAI in the capacity of “head of Preparedness.” The team’s primary responsibilities encompass monitoring, forecasting, and safeguarding against various risks posed by future AI systems, ranging from their ability to deceive and manipulate humans (as seen in phishing attacks) to their potential for generating malicious code.

Preparedness is tasked with studying a range of risk categories, some of which may appear far-fetched, such as “chemical, biological, radiological, and nuclear” threats in the context of AI models. OpenAI CEO Sam Altman, known for expressing concerns about AI-related doomsday scenarios, is taking a proactive approach in preparing for such risks. The company is open to investigating both obvious and less apparent AI risks and is soliciting ideas from the community for risk studies, offering a $25,000 prize and job opportunities with the Preparedness team to top contributors.

In addition to risk assessment, the Preparedness team will work on formulating a “risk-informed development policy” to guide OpenAI’s approach to AI model evaluations, monitoring, risk mitigation, and governance structure. This approach complements OpenAI’s existing work in AI safety, focusing on both the pre- and post-model deployment phases. OpenAI acknowledges the potential benefits of highly capable AI systems but emphasizes the need to understand and establish infrastructure to ensure their safe use and operation. This announcement coincides with a major U.K. government summit on AI safety and follows OpenAI’s commitment to study and control emerging forms of “superintelligent” AI, driven by concerns about the potential for advanced AI systems to surpass human intelligence within the next decade.
You can read more details here from openAI blog.

Yuuma

Tweet

Microsoft Azure AI has unveiled a groundbreaking innovation in the realm of image development. They’ve introduced Idea2Img, a multimodal AI framework designed to simplify the process of transforming abstract concepts into tangible images, reducing the need for manual effort.

Idea2Img leverages the power of large multimodal models (LMMs) like GPT-4V to enable a self-refinement process. This iterative approach involves GPT-4V performing prompt generation, selecting draft images, and reflecting on feedback to continually improve results.

What sets Idea2Img apart is its integrated memory module, which tracks the history of exploration for each type of prompt, whether it’s a picture, text, or feedback. This constant interaction between the processes driven by GPT-4V is the key to Idea2Img’s impressive capabilities.

In practical scenarios involving intertwined picture-text sequences, visual design elements, and complex usage descriptions, Idea2Img excels. It can even extract intricate visual information from input images. To assess its effectiveness, the research team conducted user preference studies, comparing Idea2Img with various other models. The results were striking, with a remarkable 26.9% improvement when Idea2Img was paired with SDXL, underscoring its outstanding efficacy in the field.

In conclusion, Microsoft’s Idea2Img is a significant advancement in image development and design. By harnessing the potential of LMMs and iterative self-refinement, it promises to revolutionize the way we create visual assets from abstract ideas. Its adaptability in complex multimodal scenarios and substantial improvements in user preferences make it a game-changing innovation with far-reaching implications for businesses and industries reliant on image creation and design. It has the potential to enhance efficiency and output quality, ultimately leading to greater competitiveness and customer satisfaction.

Asahi

Tweet