Google’s AI Overviews Now Resemble OpenAI’s Search Prototype Amid Rising Competition
Key Highlights:
- Google has updated AI Overviews, making it resemble OpenAI’s SearchGPT prototype.
- The update introduces a right-hand link display and in-text links for easier access to websites.
- Google’s approach to publishers and its AI-powered search platform remains distinct from OpenAI’s strategy.
Google has updated its AI Overviews, intensifying the search engine competition just a month after OpenAI introduced its SearchGPT prototype. The AI Overviews, which generate answers in Google search results, now feature a right-hand display of links. Additionally, Google is experimenting with in-text links within the overviews, enabling users to view cited information directly.
The update is available on desktops and mobile devices, accessible by tapping the site icons in the upper-right corner of the screen.
Hema Budaraju, Google’s senior director of product management for Search, told Business Insider that the update aims to make it easier for users to visit websites that interest them.
Google first introduced AI Overviews in May during its annual I/O conference. The tool provides a condensed summary of search results at the top of the page for certain queries. While the earlier version did include links, they were less prominent, requiring users to scroll down or click a down arrow at the bottom of a paragraph to view the sources.
Hema Budaraju, senior director of product management for Google Search, told Business Insider that the update will make it easier for people to visit sites that interest them.
The updates make Google’s AI search product more similar to OpenAI’s SearchGPT, which also features in-text links and an expandable sidebar for sources. Powered by GPT-4, SearchGPT provides real-time web responses and is currently being tested with a limited group of users and publishers, as per the company’s announcement.
However, Google’s AI Overviews are much more widely accessible, and key differences remain. Unlike AI Overviews, SearchGPT functions as a chatbot, allowing users to ask follow-up questions within the same conversation instead of starting a new search. OpenAI also plans to integrate SearchGPT’s top features directly into ChatGPT.
In contrast, Google is embedding AI into its existing search engine, though it also offers a chatbot experience through its virtual assistant, Gemini.
The two companies also differ in their approach to deals with publishers. OpenAI has secured agreements with various news organizations, providing financial compensation in exchange for access to their content libraries to train its AI models.
These deals arise amid publisher concerns over the potential impact of AI products on readership and the unauthorized use of their content for training large language models. For instance, AI-powered search tool Perplexity faced criticism for allegedly repurposing news content without proper attribution to publishers.
While Google appears to be expanding attribution, it has not pursued similar agreements with U.S. publishers. However, Google is paying Reddit to use its vast collection of online posts for AI training. In 2023, Google also signed licensing agreements with 1,500 news publications in the EU under the EU Copyright Directive.
As companies refine generative AI search tools, publishers are closely monitoring how attribution and citation are implemented. From a user perspective, AI models are known to hallucinate, so identifying information sources can be valuable for fact-checking or further research.
Hema Budaraju stated that Google is continuing to test different ways to present information while focusing on methods that drive traffic to relevant websites. According to her, showing links directly within AI Overviews has led to increased traffic to publisher sites.
Google’s AI Overviews update coincides with its expansion to six new countries, including the UK, India, Japan, Indonesia, Mexico, and Brazil. The enhanced version with links is available in all launched regions, as well as to Search Labs users in over 120 countries and territories.
Additionally, Google is experimenting with new tools in Search Labs for U.S. English queries, such as the ability to save AI Overview results and simplify response language.
Explore the Attractive Features of AI Overviews in Search
Over the past 25 years, Google has continually evolved its Search capabilities through various technological advancements. The company has refined its core information systems to ensure users can quickly access high-quality web content, building a vast knowledge base of billions of facts about people, places, and things.
With the advent of generative AI, Google Search now offers enhanced functionality. Users can inquire about anything—from research and planning to brainstorming—and Google will handle the complexities.
This upgrade is powered by a new Gemini model specifically tailored for Google Search. The model integrates Gemini’s advanced features, including multi-step reasoning, planning, and multimodality, with Google’s premier Search systems.
Get Quick Answers with AI Overviews
Google has announced the widespread rollout of its AI Overviews feature in Search, offering users quick answers and relevant links. Previously tested in Search Labs, AI Overviews have already been used billions of times. The tool provides a brief summary of topics alongside links for further exploration, improving user satisfaction and increasing Search usage.
Starting today, AI Overviews will be available to all U.S. users, with plans to expand to more countries soon. Google expects to reach over a billion users by the end of the year. The feature is driving users to a wider range of websites for complex queries, with links in AI Overviews receiving more clicks than traditional web listings.
As Google expands this feature, it remains focused on delivering valuable traffic to publishers and creators. Ads will continue to appear in clearly marked sections, distinguishing sponsored content from organic results.
Customize Your AI Overview
Google is set to introduce new customization options for AI Overviews, allowing users to simplify or expand the language according to their needs. This feature is especially useful for those new to a topic or looking to make information more accessible for young learners.
The update will soon be available in Search Labs for English queries in the U.S.
Pose Your Most Complex Questions
The advanced Gemini model’s multi-step reasoning will enhance AI Overviews’ ability to handle complex queries. Users will be able to ask detailed questions with multiple facets in a single search.
For instance, if you’re searching for a new yoga or pilates studio that is popular with locals, conveniently located, and offers new member discounts, you’ll soon be able to request details like: “Find the best yoga or pilates studios in Boston and provide information on their intro offers and walking time from Beacon Hill.”
These multi-step reasoning features will soon be accessible in AI Overviews through Search Labs for English queries in the U.S.
Plan Ahead with Search
Google is enhancing its Search functionality with new planning capabilities, allowing users to organize and prepare for various needs, from meals to vacations. For example, you can now search for “create a 3-day meal plan for a group that’s easy to prepare” and receive a selection of recipes from across the web to get you started.
If you need adjustments, such as swapping a dinner recipe for a vegetarian option, simply make your request, and Search will tailor your meal plan accordingly. You can then easily export your customized plan to Docs or Gmail.
Meal and trip planning features are currently available in Search Labs for English users in the U.S. Later this year, Google plans to introduce additional customization options and expand into categories like parties, date nights, and workouts.
Discover an AI-Organized Results Page
Searching for inspiration can often be overwhelming, but Google’s upcoming AI-organized results page will make it easier. Generative AI will help brainstorm and categorize results under unique, AI-generated headlines, providing a variety of perspectives and content types.
This new feature will soon be available for English searches in the U.S., starting with dining and recipes, and expanding to include movies, music, books, hotels, shopping, and more.
Get Help from AI with Video Search
Google Search is evolving beyond traditional text queries. With new advancements in video understanding, users can now ask questions using video content, enhancing visual search capabilities.
For instance, if you encounter issues with a record player bought from a thrift shop—such as an erratic needle movement—searching with video will help. Instead of trying to describe the problem in words, simply upload a video, and you’ll receive an AI Overview with troubleshooting steps and resources.
Video search will soon be available to English-speaking users in the U.S. through Search Labs, with plans for expansion to additional regions over time.
This feature highlights Google’s efforts to revolutionize Search by integrating the latest advancements of the Gemini model. Soon, Google will handle searching, simplifying, researching, planning, brainstorming, and more.
Read more: