Bing Generative Search: 1. Screenshots of an AI summary from Germany – Current developments and significance

Microsoft: Bing Generative Search – The first signs in Germany

Microsoft's Bing Generative Search was first introduced in an early version in July 2024 and began pilot testing at that time. The official rollout of the feature to a wider audience in the US started in October 2024, with generative search initially available primarily for a small number of user queries and focusing on informational searches.

Regarding availability in Germany, there were no official announcements for a long time. But now, the first concrete indications suggest that the feature will also appear here: Xpert.Digital has published screenshots of Bing Generative Search in German. Although Microsoft has not yet announced an official, broad rollout for Germany or other countries outside the US, these observations suggest that Bing has already begun delivering generative search to certain users or search queries in Germany. This strongly suggests that an official launch for Germany is only a matter of time.

Milestones:

• First rollout of Bing Generative Search: July 2024 (pilot phase).
• Rollout in the USA: October 2024, gradual expansion to more users and search queries.
• Availability in Germany: No official rollout has been announced, but initial screenshots confirm the feature is beginning to be rolled out. An official launch is expected soon.

Microsoft apparently plans to gradually roll out the feature to other regions after further testing and feedback. Given the screenshots that have now surfaced from Germany, an official launch here is likely soon, although specific dates are not yet known.

Microsoft Bing Generative Search: The Transformation of Search Engines

Microsoft has taken a significant step toward AI-powered web search with Bing Generative Search. This innovative feature transforms the traditional search engine from a simple collection of links into a dynamic, magazine-like experience that presents information directly on the search results page and provides contextual answers. The technology is based on a combination of large and small language models (LLMs and SLMs) as well as Microsoft's proprietary Prometheus technology, which combines the Bing search index with AI-powered reasoning. Following an initial testing phase starting in July 2024, the feature is now being rolled out to more US users, initially focusing on informational search queries. The first appearances now observed in Germany indicate that the global expansion has already begun. Microsoft emphasizes that the feature was designed to maintain traffic to the original sources, thus promoting a healthy web ecosystem.

Related to this:

• Between 67% and 90% | B2B prefers web searches using AI tools instead of traditional search engines

What is Bing Generative Search?

Bing Generative Search represents a fundamental reimagining of the search engine. Instead of simply displaying links to relevant websites, the system generates a tailored and dynamic response to the user's query. This new experience combines the foundation of Bing's search results with the creativity and reasoning capabilities of modern language models.

For a search query like “What is a Spaghetti Western?”, Bing generates an AI-powered answer that delves deep into this film genre, explaining its history and origins, and providing well-known examples. The information is presented in an easy-to-understand format, displaying links and sources that verify the information's origin or allow users to explore the topic further. Traditional search results remain prominently displayed on the page, allowing users to choose between the AI-generated summary and the classic web links.

Microsoft describes this new experience as a transformation of the traditional Bing search results page into a more engaging, magazine-like experience that is both informative and visually appealing. The presentation incorporates rich, contextual information, including text, images, and interactive elements, providing a seamless learning experience.

The evolution from web search to generative AI

Bing Generative Search marks the beginning of a transformation in online search. This method integrates artificial intelligence with search engine optimization – a concept that will likely be further refined. Microsoft emphasizes that they are rolling out the feature “slowly to ensure we deliver a quality experience before making it generally available.” The first signs now observed in Germany indicate that this gradual rollout is also beginning there.

Bing Generative Search results appear in an organized layout that optimizes aesthetic appeal. This improved appearance also makes it easier to search for and find new information, as various topics are divided into categories along with images. While the feature already demonstrates impressive capabilities, it is still in an early stage of development and has not yet reached its full potential.

Related to this:

• Development of Microsoft search engines and their predecessors before Bing

The technological architecture behind Bing Generative Search

The Prometheus technology

The technology powering Bing Generative Search is based on Microsoft's proprietary Prometheus model. This AI model combines Bing's current and comprehensive index, ranking, and answers with the creative reasoning capabilities of OpenAI's most advanced GPT models.

Prometheus technology works in several key areas:

• Query interpretation: It takes a long, spoken query and breaks it down into a compact, search engine-friendly query so that Bing Chat can process it and find content faster.
• Bing Index: It uses Bing's search index so that Bing Chat can use up-to-date information in real time.
• Bing ranking: Bing's ranking algorithm determines which content should be displayed in the answer and which documents ChatGPT should use for the answers.
• Bing responses and results: Bing can display responses such as weather, sports results, message boxes, local results and/or even advertisements directly in Bing chat responses.
• Quotes and links: Bing Chat offers links and quotes to the sources of the content, a functionality that Microsoft says is only possible through Prometheus technology.

The integration of language models

Bing Generative Search uses a combination of large language models (LLMs) and small language models (SLMs). The LLM understands the search query, checks millions of information sources, dynamically matches content, and generates results in a new, engaging layout to quickly help the user find the answer they're looking for.

In developing Bing Generative Search, Microsoft refined various methods to optimize accuracy and incorporated these insights into the further development of the use of LLMs in search.

Functionality and user experience

The new user interface

Bing Generative Search's user interface differs significantly from that of traditional search engines. It offers a document directory that facilitates navigation to related sections and enhances the user experience by making information more accessible. Organic search results are positioned to the right of the answer summary, ensuring they remain visible and relevant. Furthermore, source citations are prominently displayed throughout the text.

The page contains the following AI and non-AI elements:

• Answer summary
• Document index
• Sources
• Related sections
• Traditional search results

Examples and use cases

Bing Generative Search is particularly well-suited for informational queries and "how" questions, such as: "How can I remove background noise from my podcast recordings?" The results vary in length, depth, and detail, offering different levels of answers for users' most pressing questions.

For example, clicking on the generative search results for a query such as “Tips to make a good impression on a new customer” takes you to a full page of AI-generated information about the query, including a table of contents with sections on “Preparation Strategies”, “Effective Communication Techniques”, “Relationship Building”, and more.

Access and availability

Bing Generative Search is still in a testing or beta phase. While the official, broad rollout is currently limited to the US (as of October 2024), screenshots recently surfaced in Germany (seen by Xpert.Digital) show that Microsoft has begun rolling out the feature to individual users or search queries there as well. Exactly how and for whom the feature is currently visible in Germany (as of...) is not known in detail, as there has been no official announcement.

In the US, users could explore the feature by typing “Bing generative search” into the search bar, which displayed a carousel of demo queries. Alternatively, the “Deep search” button on the results page could be used. Similar access options are expected to be available in Germany with the official launch. However, it should be noted that the experience is still in beta. Longer loading times may occur, as Microsoft wants to ensure that generative search results are only displayed when they are reliable in terms of accuracy and relevance.

Xpert.Digital possesses in-depth knowledge across various industries. This allows us to develop tailored strategies precisely aligned with the requirements and challenges of your specific market segment. By continuously analyzing market trends and monitoring industry developments, we can act proactively and offer innovative solutions. The combination of experience and expertise generates added value and provides our clients with a decisive competitive advantage.

More information here:

• Benefit from Xpert.Digital's 5 areas of expertise in one package – starting from just €500/month

Comparison with Google AI Overviews

Microsoft's entry into AI-powered search with Bing Generative Search comes several months after Google integrated similar AI summaries into its search results. However, Google quickly faced significant criticism, as its feature spread conspiracy theories and offered dangerous health advice.

Microsoft appears to have learned from these problems and emphasizes that each of its summaries includes a source link, allowing users to verify the information and learn more. Furthermore, Microsoft initially makes the results available only for a "small percentage" of user queries. Initial observations from Germany suggest that this could also be the initial approach there.

Similar to Google's AI Overviews, Bing's AI Search also allows users to rate the AI's performance positively or negatively. However, Microsoft Bing's feedback buttons are more visible and appear at the top of the AI ​​section, while Google's AI Overviews hide their rating buttons in an expandable area that the user must click to access.

Related to this:

• Bing is important for SEO and AI search: Microsoft's growing advantage over Google in the B2B search market

Impact on the web ecosystem

Effects on publishers and traffic

Microsoft emphasizes that they are closely monitoring the impact of Bing Generative Search on publisher traffic as they develop it. Early data from the US suggests that this experience sustains website click-through rates and supports a healthy web ecosystem. It will be crucial to observe how this plays out as the feature rolls out to other regions, including Germany.

The generative search experience was developed with this goal in mind, including maintaining traditional search results while increasing the number of clickable links, such as references within the results. Microsoft ensures that additional citations and links are included, allowing users to explore further and verify accuracy, which in turn sends more traffic to publishers, thus maintaining a healthy web ecosystem.

Transparency and responsibility

Microsoft emphasizes its commitment to transparent and responsible AI. Each section in the generative AI search results lists the sources from which it draws information, allowing users to navigate directly to the original sources for even deeper insights.

Bing Generative Search results pages also include a sidebar with organic search results and suggestions for related searches, giving users the opportunity to delve even deeper into the information they are looking for and related topics.

The development of AI integration in Bing

From Bing Chat to Generative Search

The integration of AI into Bing began in February 2023, when Microsoft announced the new Bing AI search and chat experience. This marked the beginning of a significant transformation of the search engine, based on the integration of OpenAI's ChatGPT technology.

Microsoft invested billions in OpenAI to challenge Google's search dominance, which has since launched its own Google Bard AI chatbot in the testing phase. This could be the start of a new era of web search, where users tell the search engine what they are looking for in a more natural and intuitive way.

The development of Bing Generative Search builds on this earlier work and represents a natural evolution in the integration of AI into search functionality.

The technological transformation

Brad Abrams, product manager for Bing Chat and Windows Copilot, explained how the orchestration engine works with OpenAI's GPT-4 language model and with first-party and third-party tools. “We don't think of these large language models as a large repository of human information. They are, but what's interesting is the emergent behavior that arises from all this information… these models have very strong reasoning capabilities… We use this reasoning to decide which tool to invoke.”

This advanced integration of AI into search represents a significant step forward and demonstrates how Microsoft is leveraging the strengths of language models to enhance the search experience.

Related to this:

• How the Microsoft Bing search engine works: Crawling and AI at a glance – currently 16.71% market share, trending upwards

Microsoft's new approach: AI and transparency for the future of search

Microsoft Bing Generative Search represents a significant step in the evolution of web search. By integrating AI technologies such as LLMs and its proprietary Prometheus technology, Bing offers a novel search experience that goes beyond traditional link collection, delivering contextual, detailed answers directly on the search results page.

The phased rollout of this feature, starting in the US and now apparently also in Germany, demonstrates Microsoft's cautious approach, aimed at avoiding mistakes made by similar products like Google's AI summaries. By emphasizing transparency, source attribution, and maintaining traffic to publishers, Microsoft is seeking to foster a healthy web ecosystem while simultaneously leveraging the benefits of AI-powered search.

While Bing Generative Search hasn't yet officially rolled out to a broad global user base, the first observed appearances in Germany suggest that global expansion is imminent and the official launch here can't be far off. The feature will likely be further refined and expanded to include additional query types and regions, underscoring Microsoft's commitment to transforming web search through AI.

The future of web search could be fundamentally changed by these and similar innovations, although balancing AI-generated content with preserving the original web ecosystem remains a key challenge. Microsoft's approach with Bing Generative Search, which is now beginning to show its first signs in Germany, offers an interesting glimpse into this evolving landscape and could serve as a model for future developments in this area.

☑️ SME support in strategy, consulting, planning and implementation

☑️ Creation or realignment of the digital strategy and digitization

☑️ Expansion and optimization of international sales processes

☑️ Global & Digital B2B trading platforms

☑️ Pioneer Business Development

 

I would be happy to serve as your personal advisor.

You can contact me by filling out the contact form below or simply call me on +49 7348 4088 965 .

I'm looking forward to our joint project.

 

 

Xpert.Digital is a hub for industry focusing on digitalization, mechanical engineering, logistics/intralogistics and photovoltaics.

With our 360° Business Development solution, we support renowned companies from new business to after-sales.

Market intelligence, smarketing, marketing automation, content development, PR, mail campaigns, personalized social media and lead nurturing are part of our digital tools.

You can find more information at: www.xpert.digital - www.xpert.solar - www.xpert.plus