Google experiments with AI-only search as competition heats up
Google has launched an experimental search mode that replaces traditional links with AI-generated summaries, as it ramps up efforts to integrate AI amid growing competition.
Currently, Google Search displays AI Overviews, which provide summaries above traditional search results.
The new AI Mode expands on AI Overviews with improved reasoning and multimodal capabilities, allowing it to handle more complex queries, Google said in a blog post.
“You can ask anything on your mind and get a helpful AI-powered response with the ability to go further with follow-up questions and helpful web links,” the company added.
AI Mode runs on a customized version of Google’s Gemini 2.0 model, designed with advanced reasoning to handle more complex queries.
Subscribers of Google One AI Premium can access the new feature by selecting the “AI Mode” tab on the search results page, alongside options like Images and Maps.
**Addressing AI accuracy concerns**
AI Overviews has faced criticism for inaccuracies, occasionally generating misleading or incorrect information. If such issues persist, they could erode user trust and harm Google’s reputation.
“In the future, the way people search for information will shift from simple retrieval to understanding, summarizing, and presenting data in a way that is both accurate and fair,” said Thomas George, president of Cybermedia Research. “Striking this balance could make a defining moment for Google. If it fails, the company risks losing the very users and partners that helped it thrive.”
Introducing too many AI-driven changes without sufficient support for the broader web ecosystem could also trigger backlash from users and content creators.
Last month, edtech firm Chegg sued Google, claiming that AI-generated previews diminish demand for original content and make it more difficult for publishers to compete.
Google’s adaptation of the traditional search model into AI Mode is complicated by its primary monetization method – advertising that drives traffic to external sites, according to Suseel Menon, practice director at Everest Group.
“For Google, it’s a ‘search economics’ problem,” Menon said. “It’s a difficult equation for Google to solve, as generative queries cost on average 30-40% more to operate than traditional queries. Bringing AI Mode under a subscription tier addresses the long-term economics of search.”
**Competing in the AI search race**With companies like OpenAI and Perplexity rapidly advancing their AI-powered search capabilities, Google faces increasing pressure to refine its approach and stay ahead.
Data from Statcounter showed that late last year Google’s search market share dipped below 90% for the first time since 2015.
“The rivalry with companies like Microsoft and OpenAI will push Google to continually refine its strategy to keep AI-powered search relevant and profitable,” George said. “While AI-driven search enhances innovation and efficiency, it also disrupts the established digital business model, forcing stakeholders to adapt to rapid change.” Google must improve data training, human oversight, algorithms, and user feedback while ensuring accuracy, content moderation, and transparency in experimental features to stay competitive, George added.