Google's AI Overviews face significant spam problem

Google's AI Overviews, which appear at the top of search results pages, have become a significant target for spam and manipulation according to recent reports. SEO professionals are highlighting how easily the feature can be exploited, raising questions about the integrity of information presented to users.

Since their launch, AI Overviews have displayed examples of spam, misinformation, and inaccurate, biased, or incomplete results in live responses. Despite Google labeling the answers as 'experimental' and noting they 'may include mistakes,' the company has continued rapidly expanding the feature globally across search results. A recent study by SEMRush indicates AI Overviews appear on 13.14% of search results, though other industry studies have shown higher percentages.

The Fatal Flaw

SEO professionals and spammers have identified AI Overviews' critical vulnerability: the answers often regurgitate internet content verbatim or hallucinate incorrect information without proper fact-checking or corroboration from other Google sources like Business Profiles or the Knowledge Graph.

One example of this problem is incorrect business phone numbers appearing in AI Overviews, even when the correct numbers are entered in Google Business Profile. When confronted with this issue at a recent Search Central meetup in NYC, Google reportedly didn't have a solution.

Growing Manipulation Tactics

Beyond accuracy issues, AI Overviews are increasingly vulnerable to manipulation. A new type of spam is growing exponentially as more people discover its effectiveness: creating listicles that claim a person or company is the "best" in a given category, even when that article is published on the company's or individual's own website.

The tactic is remarkably simple: create an article ranking companies, include yourself as "the best," and AI Overviews will often cite you as such. Lily Ray, Vice President of SEO Strategy & Research, recently encountered several clearly AI-generated articles published within the last two months making such claims, which Google's AI Overviews then cited as sources of truth.

Ray suspects there might be "secret deals in place where companies agree to work together to cite each other in their roundup lists of 'best SEO agencies,'" explaining why the same agencies frequently appear across multiple "best SEO agency" listicles. She notes that despite 15 years in the industry and active engagement in the global SEO community, she hasn't heard of many of these supposedly top agencies, nor seen them contribute meaningfully to the field.

Google's Responsibility

The primary concern, according to Ray, lies with Google itself. "This type of spam should not be working in AI Overviews," she argues. "The fact that any of this works shows how fundamentally flawed AI Overviews are — Google is apparently not using any type of fact-checking or consensus mechanism to generate or verify its answers."

Ray finds it "mind-boggling" that Google, which pushed site owners to focus on E-E-A-T (Experience, Expertise, Authoritativeness, and Trustworthiness), is now elevating "problematic, biased and spammy answers and citations in AI Overview results." She also expresses surprise that Google is rapidly expanding AI Overviews globally despite these issues.

A Concerning Shift

The situation represents a concerning shift for Google, which was previously "considered the best in the world at identifying and neutralizing search spam and SEO manipulation," according to Ray. The company had established high standards with updates like Panda, Penguin, and core algorithm changes that targeted low-quality content.

"But with AI Overviews, it sort of feels like they've thrown that legacy to the wayside," Ray notes. "They're no longer prioritizing trustworthy, high-quality information; they're rewarding surface-level repetition and easily gamed content — at scale."

Public Implications

The broader implications are troubling given Google's massive reach. Ray questions whether it's wise to roll out potentially flawed AI features publicly "when Google receives almost 14 billion searches a day" and whether the average user will question Google's AI answers when they've trusted the company "to provide them with helpful and accurate information for 25 years."

While it's unsurprising that SEOs and spammers are exploiting the system, Ray emphasizes that "the problem is that Google is letting it happen — and in this case, they're the ones designing the system to be so easily exploitable."

Impact on Search Traffic and Publisher Revenue

The effects of AI Overviews extend beyond just presenting potentially misleading information. Recent research from Ahrefs published on April 17, 2025, directly contradicts Google CEO Sundar Pichai's claim that content and links within AI Overviews get higher clickthrough rates than content outside of them.

The comprehensive study analyzed 300,000 keywords, comparing clickthrough rates for top positions in search results with and without AI Overviews. According to Ryan Law, Director of Content Marketing at Ahrefs, the data shows that the presence of an AI Overview correlates with a 34.5% lower average clickthrough rate for the top-ranking page, compared to similar informational keywords without an AI Overview.

In a separate study from February 2025, an analysis of approximately 10,000 keywords with informational intent ranking in the top 20 positions demonstrated a significant decline in engagement metrics. According to this research, organic click-through rates for queries featuring AI Overviews dropped from 1.41% to 0.64% year-over-year, marking a 54.6% decrease.

Internal Links Further Limit External Traffic

Google has recently made additional changes that may compound these issues. On April 11, 2025, Google introduced a significant modification to its AI Overviews, implementing additional links throughout AI-generated summaries. However, these new links direct users to other Google Search pages rather than to external websites.

Ben Schoon from 9to5Google reported that users noticed AI Overviews in Google Search now contain numerous hyperlinked words and phrases. When clicked, these links don't lead to external websites but instead redirect users to additional Google Search pages related to those terms.

According to Google's statement to Search Engine Land, the company has "added links to some terms within AI Overviews when our systems determine it might be useful." Google compared this functionality to its "long-standing 'People also search for' feature," suggesting the change aims to enhance user experience by facilitating deeper exploration of topics directly within Google's ecosystem.

In one example highlighted by 9to5Google, a single AI Overview contained 31 links redirecting to other Google Search pages, compared to only 7 links pointing to external websites. This significant disparity in link distribution has raised concerns among publishers and SEO professionals about potential impacts on web traffic.

Advertising Integration

The commercial implications of AI Overviews are also expanding. In October 2024, Google unveiled an update to its search advertising platform, introducing ads within AI Overviews for mobile users in the United States. This development marked a new chapter in digital advertising, integrating promotional content into AI-generated search summaries.

AI Overviews in Google Search are designed to provide users with quick, comprehensive answers to their queries. These overviews synthesize information from various sources, offering a concise summary that saves users time and effort in piecing together information from multiple search results.

According to Shashi Thakur, VP/GM of Google Ads, the company carefully tested the integration of ads within AI Overviews over several months before the launch. The feature went live for mobile users in the United States, with potential for broader rollout in the future.

Ongoing Errors

Alongside spam issues, Google's AI Overview feature continues to experience embarrassing errors. On April 21, 2025, the system incorrectly stated that prominent SEO professional Lily Ray is 9 years old and not a dog. The AI-generated summary appeared when searching for personal information about Ray, misinterpreting information from her website about her dog Marcy, who was born in 2015, attributing these details to Ray herself.

Ray posted on X: "The 'she' on this page of my website refers to my dog, not me." The post garnered over 5,000 views and sparked discussion among SEO professionals about the AI's accuracy issues.

This incident reflects broader challenges with Google's AI Overview feature that emerged in 2024 when the system provided erroneous advice like recommending eating glue and rocks. The error gained additional attention when users discovered the AI system also failed to find Marcy the dog's age, despite it being explicitly stated on the webpage.

Timeline

• October 31, 2024: Ahrefs publishes initial research on what triggers AI Overviews, analyzing 300,000 keywords
• October 2024: Google launches ads within AI Overviews for mobile users in the US
• Late 2024: US rollout of Google's AI Overview feature begins
• February 2025: Studies show 54.6% decrease in organic click-through rates for queries with AI Overviews
• March 2025: Researchers compare clickthrough rates from before and after AI Overview implementation
• April 11, 2025: Google adds internal links within AI Overviews, directing users to other Google pages instead of external websites
• April 17, 2025: Ahrefs releases study showing 34.5% reduction in clickthrough rates when AI Overviews are present
• April 21, 2025: AI Overview incorrectly identifies SEO expert Lily Ray as a 9-year-old dog
• May 14, 2025: Lily Ray publishes detailed analysis of spam problems in AI Overviews